Home Page

SVN builds - new driver flood

Posted By: Anna Wu

SVN builds - new driver flood - 09/12/08 05:19 PM

So many new test drivers was added in the last builds.
I used the days to find/fix the missing rom sets. smile

My own status of the ROM sets

Correct : 603
Incorrect : 15
Total : 618

Now I can continue the tests.
Posted By: k1w1

Re: SVN builds - new driver flood - 09/13/08 02:18 AM

There is also a lot of double-ups. E.G.

Superpet, SP9000 & MicroMainFrame 9000 are all the same machine.

The Pet 40xx & 80xx are pretty much the same as CBM 40xx & 80xx as well.

The commodore structure tree's although better than before are still generally pretty screwy.

And the PC/AT (VGA, MF2 Keyboard) should be placed in the IBM PC/AT 5170 tree
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/13/08 02:35 AM

I don't think Judge has finished EGA yet. I'm sure VGA will get sorted in time though smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/13/08 05:06 AM

Originally Posted By k1w1
Superpet, SP9000 & MicroMainFrame 9000 are all the same machine.


It's done in purpose. There are also a bunch of similarly "duplicated" entries in c64.c. The idea is to have an entry for each different model sold (see also many Sharp Pocket Computers or Sega SC3000/SC3000h or Dina's different Coleco clones), possibly with different artworks (e.g. if we start to implement clickable keyboard we will need good scan of the different PET2001 'chicklet' and full stroke keyboards).

Originally Posted By k1w1
The Pet 40xx & 80xx are pretty much the same as CBM 40xx & 80xx as well.


Only the old models. The new ones have a different ROM with support for 60Hz / 50Hz CRTC. And models with normal keyboard vs business keyboard have different BIOSes as well (to support larger chargen in the latter ones, and numbers only on the keypad in the former ones)

Originally Posted By k1w1
The commodore structure tree's although better than before are still generally pretty screwy.


If you read the notes at the top of the driver you'll see that it reflects much better the informations from here and here.

I don't claim it's perfect, but I can ensure I spent hours to sort the previous mess (no pun intended) and the contradictory info you can find around in the web. The final goal is to make MESS source a reliable document about the original hardware (that's why there is no more CBM 20xx series, which is unconfirmed and probably fake: PET european models became CBM 30xx) and I'd like to improve it, if possible: if you find inaccuracies please be a little more precise than "are still generally pretty screwy"

Originally Posted By k1w1
And the PC/AT (VGA, MF2 Keyboard) should be placed in the IBM PC/AT 5170 tree


pc.c work is wip and only half done (see http://mess.toseciso.org/people:wilbert_pol ). newly added sets 5150, 5160 & 5170 (with clones) are actively worked on. old sets like pc, atvga and all the amstrad models are still in the previous (bad) shape, untouched. Judge will sort them out later for sure, when he's satisfied by the emulation status.
Posted By: k1w1

Re: SVN builds - new driver flood - 09/13/08 07:14 AM

etabeta78,

Thanks for explaining, it makes a bit more sense now. The Commodore product line is just so full of variants and alternate names it gets confusing.

k1w1
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/13/08 08:19 AM

SVN 3247

The max driver (Ultimax)is not working for me.
I still get a black screen. Is this ok ?
Posted By: judge

Re: SVN builds - new driver flood - 09/13/08 08:35 AM

Originally Posted By R. Belmont
I don't think Judge has finished EGA yet. I'm sure VGA will get sorted in time though smile


The EGA is locally passing all of it's POST tests. The EGA POST tests use some of the timers which are causing a timer irq some time later and that is causing the AT BIOS timer tests to break somehow.

I also have quite some updates for the DMA controller locally. The DMA locally is triggered by timer 1 to execute the RAM refreshes. I first need to get rid of the ugly DMA hack used by the pc_hdc code.


In other news: The cassette code has been changed into a mame device, this may be causing some regressions. This change also made it clear that some driver say they support cassettes but they actually simply copy the data into RAM or do some other 'evil' things. These things should hopefully be fixed before the next release. Reports of broken cassette implementations are welcome. The driver files known to have broken cassette implementations are: cgenie, intv, jupiter, nascom1, osi, ti990_10, trs80, tx0, kim1.


Posted By: etabeta78

Re: SVN builds - new driver flood - 09/13/08 08:53 AM

Originally Posted By k1w1
Thanks for explaining, it makes a bit more sense now. The Commodore product line is just so full of variants and alternate names it gets confusing.

k1w1


Reading back my answer, I could have sounded harsher than I meant. I should add more emoticons, even when I'm in hurry wink

About Commodore, they had indeed plenty of small variations among models and it's been a pain to sort them out. It could well be that a few of the duplicate sets will be removed later, if there was really no difference between them. However, if you find any mistake please report it, I could have easily overlooked or misunderstood some of the sources.

Originally Posted By Anna Wu
SVN 3247

The max driver (Ultimax)is not working for me.
I still get a black screen. Is this ok ?


Yeah. Or better, it's the best you can expect right now. It will be fixed eventually, when the emulation improves. The system at least doesn't crash when you load a cart (which was a bug reported in bugzilla), but the NOT_WORKING flag will be probably added in 0.128 if we cannot fix it before.
Posted By: mattmess

Re: SVN builds - new driver flood - 09/13/08 12:22 PM

Originally Posted By etabeta78

It's done in purpose. There are also a bunch of similarly "duplicated" entries in c64.c. The idea is to have an entry for each different model sold (see also many Sharp Pocket Computers or Sega SC3000/SC3000h or Dina's different Coleco clones), possibly with different artworks (e.g. if we start to implement clickable keyboard we will need good scan of the different PET2001 'chicklet' and full stroke keyboards).



Will the different duplicate Master Systems be added? I remember a driver for the Master System III Compact (the first system I owned), but it was removed =(.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/13/08 03:13 PM

For systems where they're electrically identical but the case is different I would recommend making the clones different views in the layout file rather than cluttering up MESS with useless duplicate drivers.
Posted By: Heihachi_73

Re: SVN builds - new driver flood - 09/13/08 03:31 PM

Speaking of Master Systems, can someone test Bank Panic on their end? It seems to be broken on all versions except smsj. CRC of the tested ROM image is B4DFB825; works with Meka's built-in BIOS (unless it's just a 'watch the SEGA logo' before the game starts), and in Dega.

Tested in MESS 0.126 with the v1.3, Hang-on and Alex Kidd BIOS sets, both NTSC and PAL.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/13/08 04:41 PM

Maybe it's a Japanese-region cart? Non-MESS emulators tend to skip region checks.
Posted By: mattmess

Re: SVN builds - new driver flood - 09/13/08 05:41 PM

Bank Panic in Japan only for SG-1000.
The Master System cartridge was never released there, AFAIK.

And speaking of Master III again... Aren't the Masters from Tec Toy different from the others? They use PAL-M, not PAL. The games run faster on PAL-M.
Posted By: Heihachi_73

Re: SVN builds - new driver flood - 09/13/08 06:48 PM

Looks like the version I've had for years is a bad dump. The [a1] set works fine. Old set is now exterminated!

Update (prior to posting) - Strangely though, the only differences in both ROMs were the last two bytes at 0x7FFE - 2020 (old version) versus 004C (a1 version).
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/08 08:53 AM

SVN 3254
Windows binary
GUI
Properties - Miscellaneous

It seems one checkbox is superimposed .


Posted By: judge

Re: SVN builds - new driver flood - 09/14/08 09:04 AM

Originally Posted By Heihachi_73
Speaking of Master Systems, can someone test Bank Panic on their end? It seems to be broken on all versions except smsj. CRC of the tested ROM image is B4DFB825; works with Meka's built-in BIOS (unless it's just a 'watch the SEGA logo' before the game starts), and in Dega.

Tested in MESS 0.126 with the v1.3, Hang-on and Alex Kidd BIOS sets, both NTSC and PAL.


Together with your next post this sounds like a checksum issue. The japanese bioses never calculate and match the checksums reported by a cartridge. The old sg1000 games did not have the checksum byte in their headers, so to keep backwards compatible sega removed the checksum checking from the japanese bios code. The euro and us bioses do calcuate and check the checksum however.

So, the sg1000 image that works on a euro/us bios is actually a checksum-fixed version of the original rom image.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/15/08 08:34 AM

SVN 3266

Othello Multivision driver

ROM check : not found
Maybe need change to : None required like for SG-1000

Othello Multivision cartridges are working with omv driver.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/15/08 08:53 AM

No, as you can see in the source, a BIOS is possibly missing because no omv unit has been dumped yet. The message is correct.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/15/08 08:59 AM

r3261
[JUPITER] Fixed regression; driver works again.

I can load game cartridges (.ace), but i am not able after loading to start/control with the keyboard (Joystick is not available). Is this ok ?
Made a double check with the EightyOne emu, here i can start all my .ace files.

Another question :
The device view show me " Cartridge " for files with the .ace extension.
Other websites and emulators call this format "snapshot".
MESS not show me the device view for snapshots. Why ?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/15/08 09:04 AM

Originally Posted By etabeta78
No, as you can see in the source, a BIOS is possibly missing because no omv unit has been dumped yet. The message is correct.


Has the original SG-1000 a bios ? If yes, why no required compare to omv ?
Posted By: Heihachi_73

Re: SVN builds - new driver flood - 09/15/08 09:32 AM

With the backwards compatibility of the Japanese Master System, should the .sg extension be added to load those games? SG-1000 Bank Panic works fine under smsj (just renamed to .sms for now). smile
Posted By: judge

Re: SVN builds - new driver flood - 09/15/08 10:25 AM

Originally Posted By Anna Wu
r3261
[JUPITER] Fixed regression; driver works again.

I can load game cartridges (.ace), but i am not able after loading to start/control with the keyboard (Joystick is not available). Is this ok ?
Made a double check with the EightyOne emu, here i can start all my .ace files.

Another question :
The device view show me " Cartridge " for files with the .ace extension.
Other websites and emulators call this format "snapshot".
MESS not show me the device view for snapshots. Why ?


The ace files are snapshots, but they have always been implemented using a cartslot. Why, I do not know. That's just another thing that needs to be fixed.

The cassette loading is not working yet, probably because of some small emulation issue. The .tap file decoding is working because I see the header being read in, but for some reason the bios code doesn't see the end of the header block.

Thanks for checking the driver.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/15/08 10:31 AM

Originally Posted By judge
Originally Posted By Anna Wu
r3261
[JUPITER] Fixed regression; driver works again.

I can load game cartridges (.ace), but i am not able after loading to start/control with the keyboard (Joystick is not available). Is this ok ?
Made a double check with the EightyOne emu, here i can start all my .ace files.

Another question :
The device view show me " Cartridge " for files with the .ace extension.
Other websites and emulators call this format "snapshot".
MESS not show me the device view for snapshots. Why ?


The ace files are snapshots, but they have always been implemented using a cartslot. Why, I do not know. That's just another thing that needs to be fixed.

The cassette loading is not working yet, probably because of some small emulation issue. The .tap file decoding is working because I see the header being read in, but for some reason the bios code doesn't see the end of the header block.

Thanks for checking the driver.


Thank you for your answer.
I read already, that the cassette loading is not working yet.
My problem is, if i load a cartridge/snapshot, I am not able to use the keyboard.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 09/15/08 12:48 PM

Originally Posted By Anna Wu
Has the original SG-1000 a bios ? If yes, why no required compare to omv ?


SG-1000 does not have a BIOS. Othello Multivision on the other hand has a built-in Othello cartridge.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/15/08 03:52 PM

Originally Posted By Curt Coder
Originally Posted By Anna Wu
Has the original SG-1000 a bios ? If yes, why no required compare to omv ?


SG-1000 does not have a BIOS. Othello Multivision on the other hand has a built-in Othello cartridge.


Thank you for the info. Now it is clear for me.

omv : omvbios.bin (8192 bytes) - NO GOOD DUMP KNOWN
romset omv [sg1000] is best available

The external Othello cartridge has a different size (32768 bytes).
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 09:02 AM

r3272
[JUPITER] Some tape loading fixes.

Not know the actual tape loading status.
If i try to load a tape file (.wav) as example moonbuggy.wav
I get a error.

LOAD moonbuggy g

load moonbuggy
Dict: moonbuggy ERROR 10

With other tape games I get the same error.

Some tap (.wav) files will not recognized.
Example: Acevaders

Error when loading the image: Unspecified error

With EightyOne emu I can load these games.
Posted By: judge

Re: SVN builds - new driver flood - 09/16/08 11:44 AM

Originally Posted By Anna Wu
r3272
[JUPITER] Some tape loading fixes.

Not know the actual tape loading status.
If i try to load a tape file (.wav) as example moonbuggy.wav
I get a error.

LOAD moonbuggy g

load moonbuggy
Dict: moonbuggy ERROR 10

With other tape games I get the same error.

Some tap (.wav) files will not recognized.
Example: Acevaders

Error when loading the image: Unspecified error

With EightyOne emu I can load these games.


I have tried the acevaders tap file, that loaded something into memory. Looking at the wav files, they seem to be in a format which the MESS core doesn't understand (yet). A clearer error message would have been nice, I know.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 11:50 AM

Originally Posted By Anna Wu
So many new test drivers was added in the last builds.
I used the days to find/fix the missing rom sets. smile

My own status of the ROM sets

Correct : 603
Incorrect : 15
Total : 618

Now I can continue the tests.


SVN revision 3273

Correct : 617
Incorrect : 11
Total : 628

I have a question to the Mess -verifyroms

vip : chip8.bin (512 bytes) - INCORRECT CHECKSUM:
EXPECTED: CRC(438ec5d5) SHA1(8aa634c239004ff041c9adbf9144bd315ab5fc77)
FOUND: CRC(5e035c8e)
romset vip is bad

mekd2 : jbug.rom (1024 bytes) - INCORRECT CHECKSUM:
EXPECTED: CRC(a2a56502) SHA1(60b6e48f35fe4899e29166641bac3e81e3b9d220)
FOUND: CRC(5ed08792)
romset mekd2 is bad

cybikov1: cyrom112.bin (32768 bytes) - INCORRECT CHECKSUM:
EXPECTED: CRC(9e1f1a0f) SHA1(6fc08de6b2c67d884ec78f748e4a4bad27ee8045)
FOUND: CRC(268da7bf)
romset cybikov1 is bad

cybikoxt: cyrom150.bin (32768 bytes) - INCORRECT CHECKSUM:
EXPECTED: CRC(18b9b21f) SHA1(28868d6174eb198a6cec6c3c70b6e494517229b9)
FOUND: CRC(268da7bf)
romset cybikoxt [cybikov1] is bad

comx35p : 80.column.card.u3 (2048 bytes) - INCORRECT LENGTH: 17 bytes
comx35p : mc6845.font.u (2048 bytes) - NO GOOD DUMP KNOWN
romset comx35p is bad
comx35n : 80.column.card.u3 (2048 bytes) - INCORRECT LENGTH: 17 bytes
comx35n : mc6845.font.u (2048 bytes) - NO GOOD DUMP KNOWN
romset comx35n [comx35p] is bad

newbrain: cop420.419 (1024 bytes) - INCORRECT LENGTH: 8192 bytes
newbrain: 8248r7.453 (4096 bytes) - NOT FOUND - NO GOOD DUMP KNOWN
romset newbrain is bad

newbraim: cop420.419 (1024 bytes) - INCORRECT LENGTH: 8192 bytes
newbraim: 8248r7.453 (4096 bytes) - NOT FOUND - NO GOOD DUMP KNOWN
romset newbraim is bad

newbraia: cop420.419 (1024 bytes) - INCORRECT LENGTH: 8192 bytes
newbraia: 8248r7.453 (4096 bytes) - NOT FOUND - NO GOOD DUMP KNOWN
romset newbraia [newbrain] is bad

newbraiv: cop420.419 (1024 bytes) - INCORRECT LENGTH: 8192 bytes
newbraiv: 8248r7.453 (4096 bytes) - NOT FOUND - NO GOOD DUMP KNOWN
romset newbraiv [newbrain] is bad

If for these drivers no (good) dumps exist, is it no better to show as best available rom sets ?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 12:54 PM

SVN revision 3273

VIC 20/1001
Not have problems to load PAL/NTSC cartridges.
Also no more have multisession problems. But not sure, maybe i mistake it with other Commodore drivers.

C16/Plus4
The cartridge Jack Attack.bin is not working.

Plus4
On C16 emu i can load and start the Jack Attack snapshot file (.prg)
On Plus 4 i can load but not start the same file.

CBM30xx/40xx/80xx
Some drivers are marked as working on GUI. But during the initializing these drivers freeze.
Posted By: Justin

Re: SVN builds - new driver flood - 09/16/08 02:49 PM

All those -verifyroms errors indicate that you need different files. You're right that the NO GOOD DUMP KNOWN stuff is not a problem, but you'll notice that each of those sets also has an INCORRECT LENGTH error on a different file - that's why it's saying they're bad.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 05:44 PM

Originally Posted By Justin
All those -verifyroms errors indicate that you need different files. You're right that the NO GOOD DUMP KNOWN stuff is not a problem, but you'll notice that each of those sets also has an INCORRECT LENGTH error on a different file - that's why it's saying they're bad.


I know some ROM files have INCORRECT LENGTH.
So i have to use fake files with needed lenght to outsmart the verify option.
Change the
newbrain: cop420.419 (1024 bytes) - INCORRECT LENGTH: 8192 bytes
newbrain: 8248r7.453 (4096 bytes) - NOT FOUND - NO GOOD DUMP KNOWN
romset newbrain is bad
to
newbrain: cop420.419 (1024 bytes) - NO GOOD DUMP KNOWN
newbrain: 8248r7.453 (4096 bytes) - NOT FOUND - NO GOOD DUMP KNOWN
romset newbrain is best available

If no dumps with correct file size exist, will be this not the same if i use a fake file with right file size or not ?
The rom set is best available.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/16/08 06:08 PM

You shouldn't use fake files. The audit doesn't count undumped and no-good-dump ROMs as bad for your final tally, as far as I know. (And ClrMAME definitely doesn't).
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 06:37 PM

Originally Posted By R. Belmont
You shouldn't use fake files. The audit doesn't count undumped and no-good-dump ROMs as bad for your final tally, as far as I know. (And ClrMAME definitely doesn't).


OK, boss. smile I listen you.


Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 06:41 PM

The Apple 1 emulator seems to be ok.
Today, was my first time to load and play a game (Lunar Lander).
Not easy to load because must use and write the right memory regions smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/16/08 06:42 PM

Wanna post how you did it so everyone can benefit? smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 06:50 PM

First you need to know which memory region the game need.
I use a machine-language game ( not basic ) called Lunar Lander with .wav extension which use the memory region " 300.A00R ".

Start Apple 1 emulator
Type " C100R "
Return
Type " 300.A00R "
Return
Play Tape
Stop Tape
Type " 300R "
Return

C100R is the command to start the cassette interface's mini-monitor.
As I said, the memory region address can be different from software to software.
It can be also, it exist the same software titles which use different addresses.
However, if you find any software in the internet, you will find also the note for
the memory region.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 08:55 PM

The Apple I Owners Club (applefritter) show me link for Lunar Lander.
Quote:
Applefritter is not a place to discuss software piracy or the distribution of copyrighted software

So i think this game is not copyrighted software.

You can also search for " iceandfire1 " and you will find the game Lunar Lander in MP3 format. Convert the MP3 to WAV.

A direct link I will not post.
Posted By: judge

Re: SVN builds - new driver flood - 09/16/08 09:08 PM

Cool to hear that it loads on the apple1 driver smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/08 09:21 PM

Here a screenshot after loading.

Posted By: etabeta78

Re: SVN builds - new driver flood - 09/16/08 10:01 PM

yep, you can also find a link to another lunar.wav file to be loaded in a different memory location. that one works as well.

unfortunately, the basic.wav coming with this second lunar.wav is not loaded correctly frown

I will try the mp3 one, and see if it works better (once converted to wav)

however, thanks Anna Wu, I would have never found those files without your successful report smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/17/08 12:21 AM

SVN revision 3273

Laser 2xxx

I am not able to load any tape file (.wav)
It can not detect the end of tape file and is still waiting ...
No problems to load tapes (.wav) with VZEM and JVZ200 emulators.
Posted By: That70sguy

Re: SVN builds - new driver flood - 09/17/08 01:02 AM

Dose anybody know, What other software titles, are avalible for the Apple 1 confused??
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 09/17/08 02:28 AM

Originally Posted By Curt Coder
Originally Posted By Anna Wu
Has the original SG-1000 a bios ? If yes, why no required compare to omv ?


SG-1000 does not have a BIOS. Othello Multivision on the other hand has a built-in Othello cartridge.


Yes, and I lost an auction for an Othello Multivision awhile back. Perhaps one will show up again someday.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 09/17/08 10:41 AM

Originally Posted By Anna Wu
omv : omvbios.bin (8192 bytes) - NO GOOD DUMP KNOWN
romset omv [sg1000] is best available

The external Othello cartridge has a different size (32768 bytes).


The 8192 bytes is just a guess, I don't know the real size of the Othello Multivision BIOS.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/17/08 06:32 PM

Originally Posted By Curt Coder
Originally Posted By Anna Wu
omv : omvbios.bin (8192 bytes) - NO GOOD DUMP KNOWN
romset omv [sg1000] is best available

The external Othello cartridge has a different size (32768 bytes).


The 8192 bytes is just a guess, I don't know the real size of the Othello Multivision BIOS.


It seems, 2 different types of Othello Multivision exist.

FG-1000 and FG-2000
Posted By: Justin

Re: SVN builds - new driver flood - 09/18/08 02:08 AM

IMO there's not much point in having a driver enabled if we have no dump of it at all....
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 09/18/08 02:30 AM

Originally Posted By Justin
IMO there's not much point in having a driver enabled if we have no dump of it at all....


That's a good point. As far as I know, it plays carts identically to the SG-1000, so the only unique point is the Othello game we don't have a dump of. Maybe just a comment in the driver would suffice for now.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 09/18/08 07:47 AM

Originally Posted By ranger_lennier
Originally Posted By Justin
IMO there's not much point in having a driver enabled if we have no dump of it at all....


That's a good point. As far as I know, it plays carts identically to the SG-1000, so the only unique point is the Othello game we don't have a dump of. Maybe just a comment in the driver would suffice for now.


That and the keyboard attached to the unit...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/08 08:03 AM

SVN revision 3283

Adventure Vision

Turtles looks a bit better but not sure it display the graphic correct.
The original Entex Adventure Vision used red LED`s with also flicker effects.
Unfortunately, i can not compare the screen because it seems no other emulator exist for this system.
Can only compare with an system which use the similar technology, Virtual Boy, but maybe make no sense.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/08 08:59 AM

SVN revision 3283

Apple3 emulation

Looks not bad but not finished yet.

Not easy to find the right software for this system.
Search for : Apple3 SOS
and you will find some in .do format.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/08 10:02 AM

SVN revision 3283

Nascom 1/Nascom 2

I am not able to load and run snapshots (.nas) on Nascom 1 emulation.
With Nascom 2 emulation I have no problems.

Also can it be, the tape emulation for both drivers is not finish yet ?
Because the tape control menu is not available.
Posted By: judge

Re: SVN builds - new driver flood - 09/18/08 10:11 AM

The tape emulation for at least nascom1 is broken by the cassette device update.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/18/08 10:25 AM

it has also never worked as a real tape: it loaded bits directly from CAS to memory frown

however, the cassette device update had the nice side effect to bring up these regressions which will eventually be fixed.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/08 10:32 AM

I have also a problem on windows gui.
Can not run
"-bios 1" or "-bios T1" = NASBUG1.T1 (original version)
"-bios 2" or "-bios T2" = NASBUG1.T2 (1978)
Even I select in Properties - Miscellanous - Bios - NasBug T1 or Nasbug T2
If i start the emulation it show me always Nasbug T4 (NASBUG 4).
The only way to start the right Bios is to run on command line.
As example: mess.exe nascom1 -bios T2

Posted By: Firewave

Re: SVN builds - new driver flood - 09/18/08 10:48 AM

Originally Posted By Justin
IMO there's not much point in having a driver enabled if we have no dump of it at all....


But with the core's ever-changing structures and validations we will be able to fix those when they are introduced and not when we somehow got a dump.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/08 11:49 AM

SVN revision 3283

Microbee 56

Can not load (boot)disk images, get a crash.
Tested with gui and on command line.
With Nanowasp emulator no problems.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/08 05:27 PM

Some screenshots (not made by MESS) about old and new drivers for Eastern European Computers and Clones.
Some drivers are very new and not working at moment. So I not make shots.
Posted By: judge

Re: SVN builds - new driver flood - 09/18/08 06:34 PM

Originally Posted By Anna Wu
SVN revision 3283

Nascom 1/Nascom 2

I am not able to load and run snapshots (.nas) on Nascom 1 emulation.
With Nascom 2 emulation I have no problems.

Also can it be, the tape emulation for both drivers is not finish yet ?
Because the tape control menu is not available.


I just took a quick look at the nascom drivers. The cassette port seems to be connected to a part (a 6402 uart) which is not hooked up in the driver code. It looks like this may require a bit more work to get cassette support working again.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/08 08:42 PM

Originally Posted By etabeta78
yep, you can also find a link to another lunar.wav file to be loaded in a different memory location. that one works as well.

unfortunately, the basic.wav coming with this second lunar.wav is not loaded correctly frown

I will try the mp3 one, and see if it works better (once converted to wav)

however, thanks Anna Wu, I would have never found those files without your successful report smile


I agree. It seems other emulators have the problem too.
Also tried to load the basic tape according to the original Apple Basic manual without success.

Posted By: F. Valero

Re: SVN builds - new driver flood - 09/19/08 10:21 AM

About the Apple 3 software you could find some in this link:

[censored]

The server is from Apple Computer and should be right and legal. (No it's not - RB)
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 09/19/08 10:44 AM

That's not an Apple server. [censored] is a user group FTP archive containing "abandonware" - software no longer sold, supported or maintained but still under copyright. It's just like a MAME ROM site.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 09/19/08 11:42 AM

Originally Posted By Anna Wu
I have also a problem on windows gui.
Can not run
"-bios 1" or "-bios T1" = NASBUG1.T1 (original version)
"-bios 2" or "-bios T2" = NASBUG1.T2 (1978)
Even I select in Properties - Miscellanous - Bios - NasBug T1 or Nasbug T2
If i start the emulation it show me always Nasbug T4 (NASBUG 4).
The only way to start the right Bios is to run on command line.
As example: mess.exe nascom1 -bios T2


This is because MAMEUI writes the full descriptive name of the BIOS to <system>.ini instead of the short name.

#
# CORE MISC OPTIONS
#
bios "F&M BASIC V3.1 with Expansion Box"


should be


bios fmbasic31
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/08 11:44 AM

Originally Posted By Vas Crabb
That's not an Apple server. [censored] is a user group FTP archive containing "abandonware" - software no longer sold, supported or maintained but still under copyright. It's just like a MAME ROM site.


A reason why i not post a direct link.

Ups ... smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/08 12:07 PM

Originally Posted By Curt Coder
Originally Posted By Anna Wu
I have also a problem on windows gui.
Can not run
"-bios 1" or "-bios T1" = NASBUG1.T1 (original version)
"-bios 2" or "-bios T2" = NASBUG1.T2 (1978)
Even I select in Properties - Miscellanous - Bios - NasBug T1 or Nasbug T2
If i start the emulation it show me always Nasbug T4 (NASBUG 4).
The only way to start the right Bios is to run on command line.
As example: mess.exe nascom1 -bios T2


This is because MAMEUI writes the full descriptive name of the BIOS to <system>.ini instead of the short name.

#
# CORE MISC OPTIONS
#
bios "F&M BASIC V3.1 with Expansion Box"


should be


bios fmbasic31



Thank you for the info.
I add now following in the nascom1.ini file :

#
# CORE MISC OPTIONS
#
bios T1

Now is ok, can load NasBug T1 bios with gui.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/19/08 12:24 PM

Posting the complete name of a site is just as bad as posting an actual working link. Please do not do either, thanks.

And for the record, "official Apple sites" end in apple.com smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/08 06:04 PM

Quote:
r3291
BIOS selection from MESSUI is now possible. (bug #1681)


Thank you, Curt. Will test it tomorrow. smile
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 09/20/08 12:15 AM

Originally Posted By R. Belmont
Posting the complete name of a site is just as bad as posting an actual working link. Please do not do either, thanks.


Sorry, too much beer and wine, and it was pirate day, after all wink
Posted By: F. Valero

Re: SVN builds - new driver flood - 09/21/08 09:45 AM

I'm Very, very sorry, I write in a hurry before leaving my work late, and write it from memory, I never want to break the rules of the board.

If you see my history of postings I have never do that and it's not going to happen again.

Again, I apologize.
Posted By: judge

Re: SVN builds - new driver flood - 09/24/08 02:30 PM

The KIM-1 driver has been updated to use clickable inputs and the emulation can now read back it's own written cassette images. I have not yet found any direct wav dumps of KIM-1 cassettes, so I have not been able to test those.

If you want to try the KIM-1 emulation be sure to start with the debugger enabled and manually step through one or two cycles before the artwork is displayed. This is a mame core issue.

The KIM-1 sysinfo page on the mess wiki have been updated with instructions on how to write to and read from tape.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/24/08 03:21 PM

Originally Posted By Anna Wu
Quote:
r3291
BIOS selection from MESSUI is now possible. (bug #1681)


Thank you, Curt. Will test it tomorrow. smile


It works well. smile

Originally Posted By judge
The KIM-1 driver has been updated to use clickable inputs and the emulation can now read back it's own written cassette images. I have not yet found any direct wav dumps of KIM-1 cassettes, so I have not been able to test those.

If you want to try the KIM-1 emulation be sure to start with the debugger enabled and manually step through one or two cycles before the artwork is displayed. This is a mame core issue.

The KIM-1 sysinfo page on the mess wiki have been updated with instructions on how to write to and read from tape.



Will test tomorrow. Need SVN higher than revision 3315.
Because if i try to load KIM-1 driver on my current version i get a crash.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/24/08 04:22 PM

Originally Posted By judge
The KIM-1 driver has been updated to use clickable inputs and the emulation can now read back it's own written cassette images. I have not yet found any direct wav dumps of KIM-1 cassettes, so I have not been able to test those.

If you want to try the KIM-1 emulation be sure to start with the debugger enabled and manually step through one or two cycles before the artwork is displayed. This is a mame core issue.

The KIM-1 sysinfo page on the mess wiki have been updated with instructions on how to write to and read from tape.



KIM-1

Not sure it is helpful.

It exist a tool where you can converting a
program file to an audio wavefile and reverse.
I convert the qchess.bin to qchess.wav
kimtape -M -A0200 -D01 -B2 qchess.bin qchess.wav
See the parameters (link).
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 07:30 AM

SVN revision 3326

KIM-1 driver

It seems it exist a problem by using the Artwork.
If I am using Artwork, i get a crash (windows gui + commandline)
Be all the same, to try start the emulation or call properties.
Only without using Artwork, I can start emulation.
But still get a crash if I call properties in windows gui-

This problem not exist on last official version 0.127.
So I think my Artwork is ok.
Posted By: judge

Re: SVN builds - new driver flood - 09/25/08 07:34 AM

Interesting. On OSX/sdlmess it starts fine, but when I change the View from 'Full' to 'Cropped' in the video options I also get a crash. Maybe our crashes are related?

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 07:37 AM

Originally Posted By judge
Interesting. On OSX/sdlmess it starts fine, but when I change the View from 'Full' to 'Cropped' in the video options I also get a crash. Maybe our crashes are related?



Maybe is related but on last official version it is working for me.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 07:57 AM

SVN revision 3326

Max driver

Great, my Commodore Ultimax cartridges are working now.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/25/08 08:38 AM

the problem with kim1 artwork is that Bobz only distributes the exe, so you have an outdated artwork. the crash is expected.

regarding max, yeah I fixed it yesterday smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 08:56 AM

Originally Posted By etabeta78
the problem with kim1 artwork is that Bobz only distributes the exe, so you have an outdated artwork. the crash is expected.

regarding max, yeah I fixed it yesterday smile


OK; will check the SVN.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 09:39 AM

DL new KIM-1 Artwork from SVN !
Delete .ini + .cfg files.

SVN revision 3326

KIM-1 driver
New Artwork show me not all.



Official version 0.127

KIM-1 driver
Old Artwork








Posted By: judge

Re: SVN builds - new driver flood - 09/25/08 09:41 AM

Did you start kim1 driver with the debugger enabled or disabled?

For now it needs to be enabled and you need to single step one or two instructions before the artwork gets displayed.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 09:53 AM

Originally Posted By judge
Did you start kim1 driver with the debugger enabled or disabled?

For now it needs to be enabled and you need to single step one or two instructions before the artwork gets displayed.


Ups, not easy.

SVN revision 3326

KIM-1 driver

New Artwork



PS: Exist a way to start the debugger by using windows gui ?
Posted By: judge

Re: SVN builds - new driver flood - 09/25/08 10:03 AM

Once u5 is out and merged it should work normally without using the debugger.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 10:28 AM

Originally Posted By judge
Once u5 is out and merged it should work normally without using the debugger.


Not have. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/08 10:43 AM

Found the option in windows gui to enable debugger.
Also can add/edit the .ini file by hand.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/28/08 06:09 PM

Originally Posted By judge
Once u5 is out and merged it should work normally without using the debugger.


You are right. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/28/08 06:29 PM

Originally Posted By Anna Wu
SVN revision 3283

Adventure Vision

Turtles looks a bit better but not sure it display the graphic correct.
The original Entex Adventure Vision used red LED`s with also flicker effects.
Unfortunately, i can not compare the screen because it seems no other emulator exist for this system.
Can only compare with an system which use the similar technology, Virtual Boy, but maybe make no sense.


SVN revision 3338

Can it be some problems with the driver exist ?
Have problems to view Turtles on screen.
Revision 3283 was much better.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 10:01 AM

SVN revision 3349

Colour Genie driver

I have problems tu run disks.
Is it correct to remove the first 450 bytes of the header from .CGD disk and than I can rename to .DSK ?
Because if i mount the disk and start emulator, I get a crash.

Jürgen Buchmüller's MESS page is no more available to get some disk´s in right format.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 10:05 AM

SVN revision 3349
Windows binary

Mac drivers

Can it be the MAC drivers are broken ?
As example the macplus driver.
If i try start the screen switch to black screen and keep it.
No problems with official 0.127 version.
Posted By: judge

Re: SVN builds - new driver flood - 09/30/08 10:59 AM

Originally Posted By Anna Wu
SVN revision 3349

Colour Genie driver

I have problems tu run disks.
Is it correct to remove the first 450 bytes of the header from .CGD disk and than I can rename to .DSK ?
Because if i mount the disk and start emulator, I get a crash.

Jürgen Buchmüller's MESS page is no more available to get some disk´s in right format.


I have no idea about disks only the cassette implementation has been improved. Maybe make a bugzilla report for the broken disk emulation.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 11:29 AM

Originally Posted By judge
Originally Posted By Anna Wu
SVN revision 3349

Colour Genie driver

I have problems tu run disks.
Is it correct to remove the first 450 bytes of the header from .CGD disk and than I can rename to .DSK ?
Because if i mount the disk and start emulator, I get a crash.

Jürgen Buchmüller's MESS page is no more available to get some disk´s in right format.


I have no idea about disks only the cassette implementation has been improved. Maybe make a bugzilla report for the broken disk emulation.


Is the cassette emulation working now ?
Convert .CAS to .WAV format (CAS2WAV)
Because
Quote:
Added proper cassette emulation. .wav only for now.

Quote:
Added preliminary .cas support. Currently only the sync signal is recognized.


Mount .WAV (basic)file
CLOAD
Play tape but not stopped
Get no prompt to type RUN.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/30/08 12:31 PM

are you sure it's a BASIC .cas and not a machine language .cas?

the former must be loaded with CLOAD the latter needs you to enter SYSTEM. see here for instructions

http://classic-computers.org.nz/system-80/driving_instructions.htm

(instructions refer to system 80, a trs80 clone, but they work as well on a color genie)
Posted By: judge

Re: SVN builds - new driver flood - 09/30/08 01:34 PM

I'm still working on fixing the internal .cas to .wav conversion for the cgenie driver. Last night the cgenie recognized the sync signal, but that was it. I need to spend some more time on that code.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/30/08 01:57 PM

Also, the Mac driver is probably broken due to the I-still-don't-understand-why changes to the 68k IRQ semantics.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 02:33 PM

Originally Posted By etabeta78
are you sure it's a BASIC .cas and not a machine language .cas?

the former must be loaded with CLOAD the latter needs you to enter SYSTEM. see here for instructions

http://classic-computers.org.nz/system-80/driving_instructions.htm

(instructions refer to system 80, a trs80 clone, but they work as well on a color genie)


Yes, the loading process for machine language based software is different (SYSTEM). I tried too, without success.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 02:35 PM

Thanks judge and Arbee for the answer.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 04:04 PM

Bridge Companion driver

Exist any plan to support the keyboard ?
I can load my cartridges but not control.

Posted By: judge

Re: SVN builds - new driver flood - 09/30/08 04:22 PM

Colour Genie .cas images now seem to load.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 04:23 PM

Originally Posted By judge
Colour Genie .cas images now seem to load.


Have to wait for Bobz next SVN release. smile

Thank You !
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/30/08 05:22 PM

SVN revision 3349

AX-350 driver
Al Alamiah MSX2 clone

It seems the driver have problems with bigger cartridge sizes.
Metal Gear 2 (512K)
Also problems to load disks.

With standard MSX2/MSX2+ driver no problems.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 07:44 AM

Originally Posted By judge
Colour Genie .cas images now seem to load.


SVN revision 3354

Colour Genie driver

The tape (.CAS) loading process looks now much better.
Still have problems, to load and start basic based software.
Some machine language based software I can load but not start or only start partially.
Maybe some compatibility problems.
I have to make more tests.

Posted By: judge

Re: SVN builds - new driver flood - 10/01/08 08:42 AM

Considering .cas loading has been broken for a couple years there are bound to be some compatibility issues, please make bug report for those issues so we cannot forget about them.

BASIC programs are still broken indeed, I do not know why yet.

Also in revision 3354 the kc and newbrain drivers are broken, they will be fixed later today.


Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 08:48 AM

Originally Posted By judge
Considering .cas loading has been broken for a couple years there are bound to be some compatibility issues, please make bug report for those issues so we cannot forget about them.

BASIC programs are still broken indeed, I do not know why yet.

Also in revision 3354 the kc and newbrain drivers are broken, they will be fixed later today.




Thank you for the info, judge. smile

The newbrain drivers may i can not test because of missing (no dump) or wrong roms.

newbrain: cop420.419 (1024 bytes) - INCORRECT LENGTH: 8192 bytes
newbrain: 8248r7.453 (4096 bytes) - NOT FOUND - NO GOOD DUMP KNOWN
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 03:09 PM

Originally Posted By judge
Considering .cas loading has been broken for a couple years there are bound to be some compatibility issues, please make bug report for those issues so we cannot forget about them.

BASIC programs are still broken indeed, I do not know why yet.

Also in revision 3354 the kc and newbrain drivers are broken, they will be fixed later today.




I have no experiences to post right bug report. frown
Here my report.
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/01/08 03:10 PM

some specific title with problems would have been preferable. wink

you can add them later, don't worry
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 03:17 PM

Originally Posted By etabeta78
some specific title with problems would have been preferable. wink

you can add them later, don't worry


For modifying the report. Which field is for the title ?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 03:33 PM

Originally Posted By Anna Wu
SVN revision 3349

AX-350 driver
Al Alamiah MSX2 clone

It seems the driver have problems with bigger cartridge sizes.
Metal Gear 2 (512K)
Also problems to load disks.

With standard MSX2/MSX2+ driver no problems.



Sorry, I made a typo.
The size of Metal Gear 2 is 524K and not 512K.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/01/08 03:37 PM

Do we know if the real Al Alamiah can use large carts? I notice it has more ROMs "built in" than most MSX2 machines so perhaps the Arabic language support or something uses that space?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 03:38 PM

Originally Posted By R. Belmont
Do we know if the real Al Alamiah can use large carts? I notice it has more ROMs "built in" than most MSX2 machines so perhaps the Arabic language support or something uses that space?


I contact Micko already about this issue. He will check. smile
262K carts working well.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 03:46 PM

Maybe I shall test the Gameboy driver only instead of exotic machines. Much easier. :))
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/01/08 03:48 PM

Testing exotic machines is valuable too - in cases where MESS is the only emulator for something it's extra important that it work properly once it's brought up smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 03:49 PM

It was more a joke. :))
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/08 04:00 PM

Micko is very helpful as developer. Especially if new drivers exist. smile
He is not overly didactic and his explanations are clear also for non programmers.

I can report him very fast,if some problems exist.

You see, I not collect roms/software for personal using only.
I try to help the developer and spend a lot of time for testing.
Unfortunately, this kind of cooperation not exist with some other developers but is ok for me.

PS: Arbee and judge are also very understandable and helpful.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/02/08 06:40 PM

It exist a nice Sonic clone game for MSX.
Unfortunately it need the V9958 graphic support.
MESS emulate this graphic chip not at all.
The game is not working.





Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/08 09:47 AM

Originally Posted By Anna Wu
SVN revision 3349

AX-350 driver
Al Alamiah MSX2 clone

It seems the driver have problems with bigger cartridge sizes.
Metal Gear 2 (524K)



Problem solved. smile Thanks, Micko !
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/08 11:04 AM

SVN revision 3379

Superboard II driver

Is the driver broken ?
If i start the driver, i have a permanent tone like playing tape.
Cannot use keyboard.

Meanwhile, I will test the WinOsi emulator.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/08 12:05 PM

SVN revision 3376

Quote:
[LYNX] Added partial support for .lyx files.


I know it is not fully working now.
Just tried to load cart with .lyx extension (converted by ucon64 tool). I can load but graphic errors.
Not sure it is depent of my conversion or of driver.

Posted By: etabeta78

Re: SVN builds - new driver flood - 10/04/08 12:23 PM

Originally Posted By Anna Wu
Superboard II driver

Is the driver broken ?


yes and no. there is some problem (the continuous sound you hear... I suggest you to turn off the sound emulation for the moment)

but to start the driver you have to press Shift + C (i.e. capital C), then press Enter twice and you're finally able to reach the command prompt.

I'll add a note to sysinfo about this.

Originally Posted By Anna Wu

Quote:
[LYNX] Added partial support for .lyx files.


I know it is not fully working now.
Just tried to load cart with .lyx extension (converted by ucon64 tool). I can load but graphic errors.
Not sure it is depent of my conversion or of driver.


the "partial support" refers to the fact that we don't support lyx images of games which use memory bank 1 (not sure if any exist however). these would probably fail to load or start both in lnx and lyx format.

the graphical issues are not related to the file format but to incomplete emulation of the lynx (and I have no idea how to improve it at the moment frown ). you would have the same exact issues with a lnx file.

thanks for testing, anyway smile
Posted By: judge

Re: SVN builds - new driver flood - 10/04/08 12:29 PM

Originally Posted By etabeta78
Originally Posted By Anna Wu
Superboard II driver

Is the driver broken ?


yes and no. there is some problem (the continuous sound you hear... I suggest you to turn off the sound emulation for the moment)

but to start the driver you have to press Shift + C (i.e. capital C), then press Enter twice and you're finally able to reach the command prompt.

I'll add a note to sysinfo about this.


This didn't entirely work for me. Looking at the todo list in driver code, a lot of work still needs to be done on this driver.

Oh, it works if I keep shift pressed.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/08 01:16 PM

Quote:


the "partial support" refers to the fact that we don't support lyx images of games which use memory bank 1 (not sure if any exist however). these would probably fail to load or start both in lnx and lyx format.

the graphical issues are not related to the file format but to incomplete emulation of the lynx (and I have no idea how to improve it at the moment frown ). you would have the same exact issues with a lnx file.

thanks for testing, anyway smile


Thanks for the info !

Lynx driver

Just a example : Blockout

Picture 1 = Blockout.lnx
Picture 2 = Blockout.lyx (ucon64 --lyx Blockout.lnx)


Posted By: etabeta78

Re: SVN builds - new driver flood - 10/04/08 01:56 PM

thanks for the example case, the 2-3 lyx files I found and tested had no issues

I'll try to find the problem and (hopefully) a solution smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/08 03:51 PM

Lynx driver
.lyx cart support

Found the first commercial game which works after conversion :
Batman Returns

Continue tests.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/08 04:31 PM

Lynx driver
.lyx cart support

It seems, noy any commercial carts works with right graphic with following cart sizes:
131K
524K

Found some commercial carts with the size of 262 which are ok.
After testing, you will get a pm.
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/04/08 06:40 PM

thanks. your remark on the cart size pointed me in the right direction.

I just submitted an update which should fix the loading of most .lyx files (still graphical issues will be present at the moment).

I keep thinking that more info on the way lynx worked internally and on the content of lynx carts would help, but empirically the new code works smile
Posted By: Justin

Re: SVN builds - new driver flood - 10/04/08 10:44 PM

There's a known regression with Lynx graphics too.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/05/08 08:31 AM

Originally Posted By etabeta78
thanks. your remark on the cart size pointed me in the right direction.

I just submitted an update which should fix the loading of most .lyx files (still graphical issues will be present at the moment).

I keep thinking that more info on the way lynx worked internally and on the content of lynx carts would help, but empirically the new code works smile


Good job, etabeta78 !

Converted 131K and 524K carts with .lyx extension seems to work now.
Maybe in future, the graphic emulation can be improved for the Lynx driver.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/05/08 08:58 AM

SVN revision 3384

TRS-80 Model 1-4 driver

Can it be, the Quickload (.cmd) option is broken ?
Quote:
Fatal error: activecpu_set_info_int() called with no active cpu! (src/emu/cpuintrf.c:1233)
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/06/08 02:02 PM

Today was my first time to split and merge roms. smile
I do it, with the new Sony MSX rom sets.

WindowsXP

1.) Merge files (roms)

As example you need to merge 2x 16K files into one 32K file:
Command line
copy 500pext.rom /b + 500pdisk.rom /b 500pext.ic47
500pext.rom 16K
500pdisk.rom 16K
500pext.ic47 32K

2.) Split files (roms)

As example you need to split one 32K file into 2x 16K files:
Start a file splitter tool like Splitfile (Freeware)
Choose the 32K file
55pbios.rom
Set the needed size > 16384 (16K)
Split the 32K file
Rename the new 2x 16K files
55pbios.ic42
55pbios.ic43
If you get a checksum error ( MESS Audition ), use no CRC in tool (without CRC you can catenate any arbitrary files).

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/06/08 07:44 PM

SVN revision 3401

Commodore max/65/c64dx driver

etabeta78, can you config the joysticks ?
By using messui and options > joysticks ... I get a crash.

AppName: messui.exe AppVer: 0.127.1.0 ModName: messlib.dll
ModVer: 0.0.0.0 Offset: 002e7b2c

Maybe for some other Commodore drivers the problem exist too.

Thank you.


Posted By: etabeta78

Re: SVN builds - new driver flood - 10/06/08 08:01 PM

press scroll lock, and then TAB and "Input (this system)"

unfortunately I have no way to test MessUI to investigate the issue.
Posted By: judge

Re: SVN builds - new driver flood - 10/06/08 08:03 PM

Maybe you need to delete an old .cfg file first.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/06/08 08:18 PM

Thanks for the answers.

I deleted already the old .cfg, was not helpful.
Quote:
press scroll lock, and then TAB and "Input (this system)"


OK, will test.
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/07/08 06:18 AM

could you also try to setup the controls from MESSGUI for genesis, nes and sms? I would be curious to see if any of them crashes as well...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/07/08 07:32 AM

Originally Posted By etabeta78
could you also try to setup the controls from MESSGUI for genesis, nes and sms? I would be curious to see if any of them crashes as well...


Can confirm the same problem with these drivers.
It seems the problem is for all driver, which use the new joystick config menu.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/08/08 08:33 AM

SVN revision 3411

snes driver

Dungeon Master (DSP-2) seems to work now.
Not know, which special chips are already supported.
Will test.
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/08/08 08:47 AM

Dungeon Master and Metal Combat are the only games which are affected by those changes. However, Metal Combat is still not working due to lack of Super Scope support, but you can see OBC-1 emulation is ok [1] because now the graphics is completely showed

It's not a huge progress in terms of supported games, but it's something we were missing previously (so it's cool to have it now) and that we got basically for free thanks to byuu's kindness to release his code as Public Domain

Apart from these two, we only support DSP-1. We miss DSP-3, DSP-4, SuperFX, SDD-1, SA-1, S-RTC, C4, SPC-7110 (both with and without RTC), ST-010, ST-011, ST-018. you can find a list of games using each chip here


[1] actually, there could still be some issues with SRAM saving/loading but it's impossible to test it without Super Scope support.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/08/08 09:15 AM

Originally Posted By etabeta78
Dungeon Master and Metal Combat are the only games which are affected by those changes. However, Metal Combat is still not working due to lack of Super Scope support, but you can see OBC-1 emulation is ok [1] because now the graphics is completely showed

It's not a huge progress in terms of supported games, but it's something we were missing previously (so it's cool to have it now) and that we got basically for free thanks to byuu's kindness to release his code as Public Domain

Apart from these two, we only support DSP-1. We miss DSP-3, DSP-4, SuperFX, SDD-1, SA-1, S-RTC, C4, SPC-7110 (both with and without RTC), ST-010, ST-011, ST-018. you can find a list of games using each chip here


[1] actually, there could still be some issues with SRAM saving/loading but it's impossible to test it without Super Scope support.


Thank you for the info !

DSP-1

Just test Super Mario Kart. Some graphical issues, not sure it is a general problem of the driver or the special chip emulation.

Picture 1 = BSNES Picture 2 = MESS

Posted By: etabeta78

Re: SVN builds - new driver flood - 10/08/08 09:33 AM

it's a general problem of the driver... that's why MESS prompts at start a red screen saying that the graphics emulation is not 100% working wink

additionally, the graphical emulation has regressed at a certain point due to some changes in the MAME core (the road in Super Mario Kart was present before, iirc).

however, all these graphical issues are covered by the starting message and no reports are necessary at the moment except if
* a game crashes or if
* a game with no special chips doesn't even start
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/08/08 09:37 AM

Daikaijuu Monogatari 2 (J)
SPC-7110

Seems to work.


Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/08/08 09:59 AM

Originally Posted By Anna Wu
Daikaijuu Monogatari 2 (J)
SPC-7110

Seems to work.




I am wrong.
The cart use the S-RTC and not the SPC-7110 chip.

Just test Momotaro Dentetsu Happy which use the SPC-7110 chip and is not working.

Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/08/08 01:28 PM

AFAIK the road in Mario Kart's always been broken (this is a timing thing rather than a DSP1 thing). Other DSP1 games like Pilotwings and Ballz3D should work fine though.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/08/08 02:26 PM

Originally Posted By R. Belmont
AFAIK the road in Mario Kart's always been broken (this is a timing thing rather than a DSP1 thing). Other DSP1 games like Pilotwings and Ballz3D should work fine though.


Thanks for the info. smile

Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 10/08/08 02:53 PM

sounds like HDMA is at least partly broken, since that is what is used to make the road 'angle away' from the screen.

Check if the tennis court floor is visible in Ace Wo Nerae, that does the same thing with hdma iirc.

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/08/08 04:40 PM

It's not HDMA (which works fine), it's a raster IRQ combined with bad cycle counts in the 65816 core. There's a reason my "todo" for SNES is basically "everything".
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/08/08 04:45 PM

the court floor is visible in Ace wo Nerae. however after a short while, the game zooms on the field while the characters become smaller and smaller and it freezes. it's like this since some time, no recent regression

EDIT: ok, after Arbee's comment no needs to check it anymore wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/09/08 08:31 AM

Originally Posted By Anna Wu
Originally Posted By etabeta78
could you also try to setup the controls from MESSGUI for genesis, nes and sms? I would be curious to see if any of them crashes as well...


Can confirm the same problem with these drivers.
It seems the problem is for all driver, which use the new joystick config menu.



SVN revision 3419

Quote:
Partial fix of newui categorized input remapping; still have some work to do.


Seems to work now. No crash, can config the joystick(s).
Not all other input devices tested.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/08 11:14 AM

SVN revision 3429

Any ABC80x driver

If I try to start the emulator, I get following error message:

src/mess/drivers/abc80x.c: abc802 CPU 1 space memory map entry references nonexistant device type Luxor ABC-77,
tag abc77:abc77
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/12/08 08:38 AM

SVN revision 3436

A7800 driver

Quote:
[A7800] Allow running without cartridge now that there's an NTSC BIOS with a built-in game


I use prototype rom : c300558-001a.bin CRC(A0E10EDF)9/15/88

Is this ok, if only the Atari 7800 PAL driver show me the internal game (Asteroids) and not the Atari 7800 NTSC driver ?




Posted By: etabeta78

Re: SVN builds - new driver flood - 10/12/08 10:45 AM

did you change the ntsc one to use the prototype bios? otherwise MESS will run the old NTSC bios with no built-in game
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/12/08 11:19 AM

Originally Posted By etabeta78
did you change the ntsc one to use the prototype bios? otherwise MESS will run the old NTSC bios with no built-in game


Thank you for te info. May I have to change the 7800.rom.
But if i make Audition (ROM check) all will passed.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/12/08 11:22 AM

SVN revision 3436

A400/A800 driver

Cartridge support

Can it be, only 16K (16.384) carts are working ?
Carts with smaller or bigger sizes are not working for me.
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/12/08 11:47 AM

Originally Posted By Anna Wu
Thank you for te info. May I have to change the 7800.rom.


no, you simply have to find the option to switch between common a7800 and the prototype BIOS

alternatively, you can run by command line

mess.exe a7800 -bios 1
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/12/08 12:18 PM

Originally Posted By etabeta78
Originally Posted By Anna Wu
Thank you for te info. May I have to change the 7800.rom.


no, you simply have to find the option to switch between common a7800 and the prototype BIOS

alternatively, you can run by command line

mess.exe a7800 -bios 1


Wow great. Thank you so much ! It´s really simple and working. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/12/08 12:56 PM

Originally Posted By Anna Wu
SVN revision 3436

A400/A800 driver

Cartridge support

Can it be, only 16K (16.384) carts are working ?
Carts with smaller or bigger sizes are not working for me.


16K working
8K, 32K, 65K, 131K not working
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/13/08 10:44 AM

Originally Posted By Anna Wu
SVN revision 3429

Any ABC80x driver

If I try to start the emulator, I get following error message:

src/mess/drivers/abc80x.c: abc802 CPU 1 space memory map entry references nonexistant device type Luxor ABC-77,
tag abc77:abc77


SVN revision 3444

The Problem seems to be fixed. Thanks Curt.

abc800m driver

I am not able to start a loaded disk.
As example: demo11.dsk

RUN DEMO

Tried the same file with ABCWIN 2.0 Beta and it works.



Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/13/08 10:51 AM

ABC80/80x drivers
It is impossible for me, to find any tape file to test the tape function on MESS. smile
Posted By: Mitch

Re: SVN builds - new driver flood - 10/13/08 04:16 PM

Originally Posted By Anna Wu
Originally Posted By etabeta78
Originally Posted By Anna Wu
Thank you for te info. May I have to change the 7800.rom.


no, you simply have to find the option to switch between common a7800 and the prototype BIOS

alternatively, you can run by command line

mess.exe a7800 -bios 1


Wow great. Thank you so much ! It´s really simple and working. smile


What's interesting about this is that the "PAL" 7800 BIOS with the built in Asteroids game is actually both NTSC and PAL. ;-) Though it was only previously enabled for PAL.
The two new 7800 BIOS prototype's that were found are NTSC only though.

Mitch
Posted By: incog

Re: SVN builds - new driver flood - 10/13/08 07:52 PM

Originally Posted By Mitch

What's interesting about this is that the "PAL" 7800 BIOS with the built in Asteroids game is actually both NTSC and PAL. ;-) Though it was only previously enabled for PAL.
The two new 7800 BIOS prototype's that were found are NTSC only though.

Mitch


Theres more than one?
Posted By: Mitch

Re: SVN builds - new driver flood - 10/13/08 08:52 PM

Originally Posted By incog
Originally Posted By Mitch

What's interesting about this is that the "PAL" 7800 BIOS with the built in Asteroids game is actually both NTSC and PAL. ;-) Though it was only previously enabled for PAL.
The two new 7800 BIOS prototype's that were found are NTSC only though.

Mitch


Theres more than one?


There are two released versions of the 7800 BIOS. The standard NTSC one with the Atari logo; and the one with the built in Asteroids game that is both NTSC and PAL but usually labeled as PAL only.

There are three prototype versions. One is an early version of the standard NTSC BIOS, one is a version of the standard NTSC BIOS and includes a built in copy of Pole Position and lastly there is the one that is an early version of the BIOS with the built in Asteroids but supports NTSC only.

There are also three newer dev BIOS releases but I don't know if you care about those.

Mitch
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/13/08 08:55 PM

Originally Posted By Mitch
Originally Posted By incog
Originally Posted By Mitch

What's interesting about this is that the "PAL" 7800 BIOS with the built in Asteroids game is actually both NTSC and PAL. ;-) Though it was only previously enabled for PAL.
The two new 7800 BIOS prototype's that were found are NTSC only though.

Mitch


Theres more than one?


There are two released versions of the 7800 BIOS. The standard NTSC one with the Atari logo; and the one with the built in Asteroids game that is both NTSC and PAL but usually labeled as PAL only.

There are three prototype versions. One is an early version of the standard NTSC BIOS, one is a version of the standard NTSC BIOS and includes a built in copy of Pole Position and lastly there is the one that is an early version of the BIOS with the built in Asteroids but supports NTSC only.

There are also three newer dev BIOS releases but I don't know if you care about those.

Mitch


You are talking about the Pole_Position_2_BIOS_Proto.bin ( 32K ), right ?
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/13/08 09:16 PM

Originally Posted By Mitch
There are three prototype versions. One is an early version of the standard NTSC BIOS, one is a version of the standard NTSC BIOS and includes a built in copy of Pole Position and lastly there is the one that is an early version of the BIOS with the built in Asteroids but supports NTSC only.

There are also three newer dev BIOS releases but I don't know if you care about those.

Mitch


yes, we care. do you have additional info and/or docs?
Posted By: incog

Re: SVN builds - new driver flood - 10/13/08 10:33 PM

Originally Posted By Mitch

There are two released versions of the 7800 BIOS. The standard NTSC one with the Atari logo; and the one with the built in Asteroids game that is both NTSC and PAL but usually labeled as PAL only.

There are three prototype versions. One is an early version of the standard NTSC BIOS, one is a version of the standard NTSC BIOS and includes a built in copy of Pole Position and lastly there is the one that is an early version of the BIOS with the built in Asteroids but supports NTSC only.

There are also three newer dev BIOS releases but I don't know if you care about those.

Mitch


Currently we only support the 2 standard NTSC and PAL dumps and the asteroids prototype bios. Do you have the checksum for the older ntsc bios version? I'm wondering if its the same as the checksum commented out in a7800.c smile
Posted By: Mitch

Re: SVN builds - new driver flood - 10/13/08 11:46 PM

Originally Posted By incog
Originally Posted By Mitch

There are two released versions of the 7800 BIOS. The standard NTSC one with the Atari logo; and the one with the built in Asteroids game that is both NTSC and PAL but usually labeled as PAL only.

There are three prototype versions. One is an early version of the standard NTSC BIOS, one is a version of the standard NTSC BIOS and includes a built in copy of Pole Position and lastly there is the one that is an early version of the BIOS with the built in Asteroids but supports NTSC only.

There are also three newer dev BIOS releases but I don't know if you care about those.

Mitch


Currently we only support the 2 standard NTSC and PAL dumps and the asteroids prototype bios. Do you have the checksum for the older ntsc bios version? I'm wondering if its the same as the checksum commented out in a7800.c smile


The one commented out is probably the old "bad" dump. I will dig them up and get the checksums posted a little later.

Mitch
Posted By: Mitch

Re: SVN builds - new driver flood - 10/14/08 04:20 AM

OK, assuming I didn't screw anything up here are the checksums:

7800 NTSC BIOS.BIN 4K FD13730C
7800 NTSC-PAL BIOS Asteroids.BIN 16K D5B61170

7800 NTSC BIOS Proto.BIN 4K 0C5E9CB5
7800 NTSC BIOS Proto Asteroids.BIN 16K A0E10EDF
7800 NTSC BIOS Proto Pole Position 2.BIN 32K FD40B838

7800 NTSC BIOS (Bad).BIN 4K 649913E5

7800 DEV BIOS 01 Eckhard Stolberg.BIN 4K 32719B3A
7800 DEV BIOS 02 Eckhard Stolberg.BIN 4K 6ED3B89A
7800 Dumper BIOS Bruce Tomlin.BIN 8K ED8CE436

Eckhard's site is: http://home.arcor.de/estolberg/tools/index.html
Bruce's site is: here

Mitch
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/08 07:50 AM

Quote:
7800 NTSC BIOS.BIN 4K FD13730C


Are you sure the CRC is right ? If I check my Bios I get 5D13730C

Thank you for posting the different bios infos !
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/08 03:12 PM

SVN revision 3448

OSCOM Nano driver

Is the driver working ?
I can mount and start a quickload file (Pollka.bin)
Can hear the sound but no any graphics.
The emulator can show any graphics ?
Posted By: Mitch

Re: SVN builds - new driver flood - 10/14/08 03:15 PM

Originally Posted By Anna Wu
Quote:
7800 NTSC BIOS.BIN 4K FD13730C


Are you sure the CRC is right ? If I check my Bios I get 5D13730C

Thank you for posting the different bios infos !


Yes, you are right it should be 5D13730C. That was a typo on my part. Sorry.

Mitch
Posted By: Curt Coder

Re: SVN builds - new driver flood - 10/14/08 06:21 PM

Originally Posted By Anna Wu
SVN revision 3448

OSCOM Nano driver

Is the driver working ?
I can mount and start a quickload file (Pollka.bin)
Can hear the sound but no any graphics.
The emulator can show any graphics ?


Press M to get to the monitor, there you can type hexadecimal digits.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/08 07:03 PM

Originally Posted By Curt Coder
Originally Posted By Anna Wu
SVN revision 3448

OSCOM Nano driver

Is the driver working ?
I can mount and start a quickload file (Pollka.bin)
Can hear the sound but no any graphics.
The emulator can show any graphics ?


Press M to get to the monitor, there you can type hexadecimal digits.


Thank you Curt ! smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/08 07:41 PM

SVN revision 3448

PC-8801 driver

I am not able to boot from Floppy Disk.

No boot problems with M88/Quasi88 emulators.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/17/08 08:34 AM

SVN revision 3458

BBC Master

I am not able to load/boot from Floppy Disk.
Tried to start emulator with original MOS 3.20 (mos3.20.rom) and enhanced MOS 3.50 ROM.

The screen show me:
Acorn MOS
This is not a language
*

Cannot use disc commands like *CAT ...

No problems with BeebEm, B-em.

Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/19/08 04:32 AM

SVN 3464 fixes the Mac not starting up problem. I've got more ADB and other work on the SE and Classic in progress but it's not ready to go in yet.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/08 08:21 AM

Originally Posted By R. Belmont
SVN 3464 fixes the Mac not starting up problem. I've got more ADB and other work on the SE and Classic in progress but it's not ready to go in yet.

Have to wait until tomorrow to get the new build for testing. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/08 03:26 PM

Originally Posted By Anna Wu
Originally Posted By R. Belmont
SVN 3464 fixes the Mac not starting up problem. I've got more ADB and other work on the SE and Classic in progress but it's not ready to go in yet.

Have to wait until tomorrow to get the new build for testing. smile


Good job, Arbee smile

Looks good for some Mac driver.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/21/08 06:50 PM

SVN revision 3478

Nathan Woods made some Windows GUI fixes already.

One small bug still exist in DIP Switches Config menu.
If the text is longer (2 lines needed) not at all is visible.



Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/29/08 03:54 PM

There's a new Atari 2600 demo out.

It comes with executables for emulators and real h/w, and both are PAL. The "real h/w" one hangs after one of the scenes in the MESS a2600p driver. The "emulator" one runs all the way through but the rotating 3d cube (yes, really) is very glitchy. (I haven't seen a video from real h/w to verify everything yet, but supposedly the latest Stella runs it correctly).

It's worth watching even for non-MESS-testing purposes incidentally, there are definitely some "how the heck did they do that on that hardware?!" moments smile
Posted By: MooglyGuy

Re: SVN builds - new driver flood - 10/29/08 06:11 PM

Originally Posted By R. Belmont
the rotating 3d cube (yes, really)


I think you meant (not really). If it rotated about anything other than the Z axis I'd be tempted to agree, but there's no reason why they couldn't just have a simple linezoom and color table in ROM. It's a bastardized road layer.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/29/08 07:27 PM

Yes, but even doing a road layer on the 2600 is a serious undertaking.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/01/08 12:22 PM

SVN revision 3513

COMX35 driver

Quote:
[COMX35] Separated 80-column ROMs driver


Dear Curt,

today I split the 80column.1.0.bin ROM (4096 bytes) which is now available in new 1802 Emulator and necessary for your added 80 Column support (80column.u3 2048 bytes + chargen.bin 2048 bytes).

I have no idea how I can test the 80 Column support because the most COMX software does not support the 80 Column card.

Do you have any tip ?

Thank you.

Posted By: Curt Coder

Re: SVN builds - new driver flood - 11/07/08 06:06 PM

Originally Posted By Anna Wu
SVN revision 3513

COMX35 driver

Quote:
[COMX35] Separated 80-column ROMs driver


Dear Curt,

today I split the 80column.1.0.bin ROM (4096 bytes) which is now available in new 1802 Emulator and necessary for your added 80 Column support (80column.u3 2048 bytes + chargen.bin 2048 bytes).

I have no idea how I can test the 80 Column support because the most COMX software does not support the 80 Column card.


You need to select the 80 column card into one of the expansion slots in the configuration, and then type CARD Sx into the emulator, where x is the number of slot the 80 column card is "in".

E.g. if 80 column card is in slot 4, type CARD S4 and press Enter.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/14/08 08:28 PM

Originally Posted By Curt Coder
Originally Posted By Anna Wu
SVN revision 3513

COMX35 driver

Quote:
[COMX35] Separated 80-column ROMs driver


Dear Curt,

today I split the 80column.1.0.bin ROM (4096 bytes) which is now available in new 1802 Emulator and necessary for your added 80 Column support (80column.u3 2048 bytes + chargen.bin 2048 bytes).

I have no idea how I can test the 80 Column support because the most COMX software does not support the 80 Column card.


You need to select the 80 column card into one of the expansion slots in the configuration, and then type CARD Sx into the emulator, where x is the number of slot the 80 column card is "in".

E.g. if 80 column card is in slot 4, type CARD S4 and press Enter.


I am back from the vacations. Thank you so much for the info, I will test it. smile

Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/08/08 07:00 PM

SVN 3771

I have a problem to audit drivers. A lot of Romsets will not detected.
Posted By: etabeta78

Re: SVN builds - new driver flood - 12/08/08 07:02 PM

from svn 3746:

Quote:
1. Updated core to MAME 0.128u5 (DISABLED MANY DRIVERS TEMPORARILY)


some have been re-enabled before 3771, other ones haven't (however, most of them have been re-enabled as of today)
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/08/08 07:08 PM

Originally Posted By etabeta78
from svn 3746:

Quote:
1. Updated core to MAME 0.128u5 (DISABLED MANY DRIVERS TEMPORARILY)


some have been re-enabled before 3771, other ones haven't (however, most of them have been re-enabled as of today)


Thank you so much for the info, etabeta78 ! smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/16/09 02:26 PM

Can it be, that in last SVN builds a lot of drivers are disabled/removed temporary ?

SVN rev. 4497

ClrMamePro

unneeded file: C:\Emulator\Mess\Bios\Al520ex.zip
unneeded file: C:\Emulator\Mess\Bios\Ax170.zip
unneeded file: C:\Emulator\Mess\Bios\Ax350.zip
unneeded file: C:\Emulator\Mess\Bios\Ax370.zip
unneeded file: C:\Emulator\Mess\Bios\Canonv10.zip
unneeded file: C:\Emulator\Mess\Bios\Canonv20.zip
unneeded file: C:\Emulator\Mess\Bios\Cf1200.zip
unneeded file: C:\Emulator\Mess\Bios\Cf2000.zip
unneeded file: C:\Emulator\Mess\Bios\cf2700.zip
unneeded file: C:\Emulator\Mess\Bios\Cf3000.zip
unneeded file: C:\Emulator\Mess\Bios\Cf3300.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc300.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc300e.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc400.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc400s.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc464.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc6128.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc6128f.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc6128s.zip
unneeded file: C:\Emulator\Mess\Bios\Cpc664.zip
unneeded file: C:\Emulator\Mess\Bios\cx5m.zip
unneeded file: C:\Emulator\Mess\Bios\Cx5m128.zip
unneeded file: C:\Emulator\Mess\Bios\Cx5m2.zip
unneeded file: C:\Emulator\Mess\Bios\Cx7m.zip
unneeded file: C:\Emulator\Mess\Bios\Cx7m128.zip
unneeded file: C:\Emulator\Mess\Bios\Dpc100.zip
unneeded file: C:\Emulator\Mess\Bios\Dpc180.zip
unneeded file: C:\Emulator\Mess\Bios\Dpc200.zip
unneeded file: C:\Emulator\Mess\Bios\Expert10.zip
unneeded file: C:\Emulator\Mess\Bios\Expert11.zip
unneeded file: C:\Emulator\Mess\Bios\Expert13.zip
unneeded file: C:\Emulator\Mess\Bios\Expert20.zip
unneeded file: C:\Emulator\Mess\Bios\Expertdp.zip
unneeded file: C:\Emulator\Mess\Bios\Expertpl.zip
unneeded file: C:\Emulator\Mess\Bios\Fs1300.zip
unneeded file: C:\Emulator\Mess\Bios\Fs4000.zip
unneeded file: C:\Emulator\Mess\Bios\Fs4500.zip
unneeded file: C:\Emulator\Mess\Bios\Fs4600.zip
unneeded file: C:\Emulator\Mess\Bios\Fs4700.zip
unneeded file: C:\Emulator\Mess\Bios\Fs5000.zip
unneeded file: C:\Emulator\Mess\Bios\Fs5500.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1a.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1f.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1fm.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1fx.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1mk2.zi
unneeded file: C:\Emulator\Mess\Bios\Fsa1wsx.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1wx.zip
unneeded file: C:\Emulator\Mess\Bios\Fsa1wxa.zip
unneeded file: C:\Emulator\Mess\Bios\Gsfc200.zip
unneeded file: C:\Emulator\Mess\Bios\Hb10p.zip
unneeded file: C:\Emulator\Mess\Bios\hb201.zip
unneeded file: C:\Emulator\Mess\Bios\hb201p.zip
unneeded file: C:\Emulator\Mess\Bios\hb20p.zip
unneeded file: C:\Emulator\Mess\Bios\Hb501p.zip
unneeded file: C:\Emulator\Mess\Bios\Hb55d.zip
unneeded file: C:\Emulator\Mess\Bios\Hb55p.zip
unneeded file: C:\Emulator\Mess\Bios\Hb75d.zip
unneeded file: C:\Emulator\Mess\Bios\hb75p.zip
unneeded file: C:\Emulator\Mess\Bios\Hbf1.zip
unneeded file: C:\Emulator\Mess\Bios\Hbf12.zip
unneeded file: C:\Emulator\Mess\Bios\hbf1xd.zip
unneeded file: C:\Emulator\Mess\Bios\Hbf1xdj.zip
unneeded file: C:\Emulator\Mess\Bios\hbf1xdm2.zip
unneeded file: C:\Emulator\Mess\Bios\Hbf1xv.zip
unneeded file: C:\Emulator\Mess\Bios\Hbf500.zip
unneeded file: C:\Emulator\Mess\Bios\hbf500p.zip
unneeded file: C:\Emulator\Mess\Bios\hbf700d.zip
unneeded file: C:\Emulator\Mess\Bios\hbf700f.zip
unneeded file: C:\Emulator\Mess\Bios\hbf700p.zip
unneeded file: C:\Emulator\Mess\Bios\hbf700s.zip
unneeded file: C:\Emulator\Mess\Bios\Hbf900.zip
unneeded file: C:\Emulator\Mess\Bios\Hbf900a.zip
unneeded file: C:\Emulator\Mess\Bios\hbf9p.zip
unneeded file: C:\Emulator\Mess\Bios\hbf9s.zip
unneeded file: C:\Emulator\Mess\Bios\Hbg900ap.zip
unneeded file: C:\Emulator\Mess\Bios\Hbg900p.zip
unneeded file: C:\Emulator\Mess\Bios\Hotbi13b.zip
unneeded file: C:\Emulator\Mess\Bios\Hotbi13p.zip
unneeded file: C:\Emulator\Mess\Bios\Hotbit11.zip
unneeded file: C:\Emulator\Mess\Bios\Hotbit12.zip
unneeded file: C:\Emulator\Mess\Bios\Hotbit20.zip
unneeded file: C:\Emulator\Mess\Bios\hx10.zip
unneeded file: C:\Emulator\Mess\Bios\Hx10s.zip
unneeded file: C:\Emulator\Mess\Bios\Hx20.zip
unneeded file: C:\Emulator\Mess\Bios\Hx23.zip
unneeded file: C:\Emulator\Mess\Bios\Hx23f.zip
unneeded file: C:\Emulator\Mess\Bios\Jvchc7gb.zip
unneeded file: C:\Emulator\Mess\Bios\Kccomp.zip
unneeded file: C:\Emulator\Mess\Bios\Mlf80.zip
unneeded file: C:\Emulator\Mess\Bios\Mlfx1.zip
unneeded file: C:\Emulator\Mess\Bios\Mpc100.zip
unneeded file: C:\Emulator\Mess\Bios\Msx.zip
unneeded file: C:\Emulator\Mess\Bios\Msx2.zip
unneeded file: C:\Emulator\Mess\Bios\Msx2p.zip
unneeded file: C:\Emulator\Mess\Bios\Nc100.zip
unneeded file: C:\Emulator\Mess\Bios\Nc200.zip
unneeded file: C:\Emulator\Mess\Bios\Nms801.zip
unneeded file: C:\Emulator\Mess\Bios\nms8220.zip
unneeded file: C:\Emulator\Mess\Bios\nms8220a.zip
unneeded file: C:\Emulator\Mess\Bios\nms8245.zip
unneeded file: C:\Emulator\Mess\Bios\nms8245f.zip
unneeded file: C:\Emulator\Mess\Bios\nms8250.zip
unneeded file: C:\Emulator\Mess\Bios\nms8255.zip
unneeded file: C:\Emulator\Mess\Bios\nms8280.zip
unneeded file: C:\Emulator\Mess\Bios\Nms8280g.zip
unneeded file: C:\Emulator\Mess\Bios\Pcw16.zip
unneeded file: C:\Emulator\Mess\Bios\Pcw8256.zip
unneeded file: C:\Emulator\Mess\Bios\Phc23.zip
unneeded file: C:\Emulator\Mess\Bios\Phc35j.zip
unneeded file: C:\Emulator\Mess\Bios\Phc70fd.zip
unneeded file: C:\Emulator\Mess\Bios\Phc70fd2.zip
unneeded file: C:\Emulator\Mess\Bios\Piopx7.zip
unneeded file: C:\Emulator\Mess\Bios\Sv328n80.zip
unneeded file: C:\Emulator\Mess\Bios\Sv328p80.zip
unneeded file: C:\Emulator\Mess\Bios\Svi318.zip
unneeded file: C:\Emulator\Mess\Bios\Svi728.zip
unneeded file: C:\Emulator\Mess\Bios\Svi738.zip
unneeded file: C:\Emulator\Mess\Bios\Svi738sw.zip
unneeded file: C:\Emulator\Mess\Bios\Tadpc200.zip
unneeded file: C:\Emulator\Mess\Bios\Tadpc20a.zip
unneeded file: C:\Emulator\Mess\Bios\Tpc310.zip
unneeded file: C:\Emulator\Mess\Bios\Vg802000.zip
unneeded file: C:\Emulator\Mess\Bios\Vg802020.zip
unneeded file: C:\Emulator\Mess\Bios\vg8230.zip
unneeded file: C:\Emulator\Mess\Bios\vg8235.zip
unneeded file: C:\Emulator\Mess\Bios\Vg8235f.zip
unneeded file: C:\Emulator\Mess\Bios\Vg8240.zip
unneeded file: C:\Emulator\Mess\Bios\Y503iir.zip
unneeded file: C:\Emulator\Mess\Bios\Y503iir2.zip
unneeded file: C:\Emulator\Mess\Bios\Yc64.zip
unneeded file: C:\Emulator\Mess\Bios\Yis303.zip
unneeded file: C:\Emulator\Mess\Bios\Yis503.zip
unneeded file: C:\Emulator\Mess\Bios\Yis503ii.zip
unneeded file: C:\Emulator\Mess\Bios\YIS503M.zip

Posted By: etabeta78

Re: SVN builds - new driver flood - 04/16/09 02:33 PM

yes, msx and amstrad (and nes in u4) have been temporarily disabled.

I think they will be back for 0.131

p.s. @Bletch: what was the main problem with these in 130u3?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/16/09 02:35 PM

Originally Posted By etabeta78
yes, msx and amstrad (and nes in u4) have been temporarily disabled.

I think they will be back for 0.131

p.s. @Bletch: what was the main problem with these in 130u3?


Ohh, thanks. I am so happy to not lost these drivers. I can wait smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/16/09 03:35 PM

Thank you my friends. smile

r4504
Fixed drivers that were disabled in the 130u3 update.
Posted By: etabeta78

Re: SVN builds - new driver flood - 04/16/09 04:03 PM

mind that amstrad computers are still disabled.

I see the problem was in the Z80 tables...
Posted By: judge

Re: SVN builds - new driver flood - 04/16/09 04:50 PM

All drivers are re-enabled in svn now. For some reason the nes drivers do not boot. I'm leaving these for someone else to look into.
Posted By: etabeta78

Re: SVN builds - new driver flood - 04/16/09 05:18 PM

Great work. If I manage to finish a small thing I'm working on, I'll take a look at NES. But be my guest if you manage to fix it smile

EDIT: Also, I was wrong... Amstrad computers were enabled in previous Judge's update. Once more, great work smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 04/17/09 10:37 AM

mmm... the NES thing is strange...

some games simply work (Legend of Zelda, Duck Tales, Mega Man 4). other games don't but video emulation starts (Startropics, SMB & SMB3). a few doesn't even start correctly video emulation (they sit on a grey screen and show the zapper crosshairs, which should disappear as soon as TIMER_CALLBACK( lightgun_tick ) is called)

I can't investigate on this any more today, unfortunately.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 04/17/09 01:32 PM

Originally Posted By etabeta78
I see the problem was in the Z80 tables...


Just a silly question: what are the Z80 cycle tables doing outside the Z80 CPU core?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 04/17/09 01:33 PM

Aaron needed to do that to emulate the super-weird Sega System 1 timing where opcode fetches are slower than operands or something like that.
Posted By: judge

Re: SVN builds - new driver flood - 04/17/09 02:25 PM

They are basically all hacks until waitstates can be properly supported in the core.
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 04/17/09 04:45 PM

I never understood why waitstates are so difficult to support.
Then again, I never really looked at the way the code works.

I sort of envision it working as an AM_WAITSTATE(cycles_to_steal) call within cpu-specific memory and i/o regions (this would be for 'simple' fixed waitstates for specific memory or i/o accesses, where the cpu clock divider and specific chip/device states are irrelevant; the ibm-pc is an example of this); and an AM_WAITSTATE_FCT(function_name) for more complicated spots which use different/optional waitstates depending on cpu mode/divider/value_written/etc.
The function would be passed the masked offset within cpu space, data for a write or read, and a cpu and device pointer(for AM_WAITSTATE_DEV?), etc. and would return the number of cycles to steal/waitstates for that particular access.
A good example of a (future)device which needs this is the PHP1500 tms5200-based speech module for the TI99/4(A), since the /READY line from the tms5200 controls the WAIT input for the main cpu; the tms5200 will halt the cpu during all reads and writes from/to it. (the exact number of cycles reads and writes take is determined by the state tables of the tms5200 and its zero or more VSM chips. This is one of those 'un-fun but inevitably needed' things to emulate.)

In addition, devices (and code in /machine/ ?) could 'force' waitstates (along with forcing cpu to catch up before waitstates are taken) using a cpu tag; this is important for the DMA chip in the ibm-pc iirc, and is also important for sprite chips which halt one or more cpus, used in all sorts of places.

This is all theoretical though, as I have NO idea how complex this would be to code. Judging by the fact that it hasn't been done, I guess its in the 'brain-bleeding' difficulty area.
Feel free to correct me if I'm wrong, though.

LN
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/20/09 06:05 AM

SVN 4525 windows build

Kaypro4 driver

If I start with windows GUI, I get a crash
Quote:
AppName: messui.exe AppVer: 0.130.1.0 ModName: messlib.dll
ModVer: 0.0.0.0 Offset: 0029f35b

Same problem on command line.

Content of kaypro4.zip:
omni_ii_u43.bin
onmi_ii_u47.bin

NGPC driver

Cannot audit/find driver.

Content of ngpc.zip:
ngp_bios.ngp
ngpcbios.rom
Posted By: Robbbert

Re: SVN builds - new driver flood - 04/20/09 07:16 AM

You have wrong roms in kaypro4.zip

Both drivers are under construction and are bound to have problems.
Posted By: judge

Re: SVN builds - new driver flood - 04/20/09 07:22 AM

The neogeo pocket driver is not enabled yet (the main cpu core is not in subversion).
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/20/09 07:27 AM

OK, i was too hasty. Thanks ! smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/20/09 07:38 AM

Originally Posted By robbbert
You have wrong roms in kaypro4.zip

Both drivers are under construction and are bound to have problems.


OK, i add :

81-146.u43
81-232.u47
Posted By: judge

Re: SVN builds - new driver flood - 04/20/09 08:43 PM

Originally Posted By judge
The neogeo pocket driver is not enabled yet (the main cpu core is not in subversion).


Have fun with the drivers smile They are enabled now.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/21/09 05:26 AM

Originally Posted By judge
Originally Posted By judge
The neogeo pocket driver is not enabled yet (the main cpu core is not in subversion).


Have fun with the drivers smile They are enabled now.


Hello judge,

just tried SVN 4532. ROM check was passed. If I start the driver (with and without mounted cart), I get a white screen.
Posted By: judge

Re: SVN builds - new driver flood - 04/21/09 05:56 AM

You have to power on the system: press Q.
I actually had to emulate the power button for this one shocked
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/21/09 06:14 AM

Thanks for your advice. It works smile smile
Posted By: judge

Re: SVN builds - new driver flood - 04/21/09 06:28 AM

At the moment you still have to enter the language and time details on every boot. On the real thing the cpu keeps running all the time at a very low frequency. The only way to do this in mess would be to play these drivers using savestates. I still have to finish the last couple of things to fully support savestates.

What already should work is boot the system, set details, power off and power on again. Between power off and power on it should be possible to change the mounted cartridge image. I haven't been able to test this fully yet though; it might be that there is some code in place which forces the emulated systems to restart when you change a mounted cartridge image.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/21/09 06:34 AM

I load the state after pressing Q.
Now i am waiting for the progress to start/play carts. smile
Posted By: judge

Re: SVN builds - new driver flood - 04/21/09 06:43 AM

You can already play the carts by just mounting them on boot.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 04/21/09 07:28 AM

Let me say, you make a great job judge !
The most games are running.
Posted By: enik

Re: SVN builds - new driver flood - 04/26/09 03:36 PM

This has been posted at bug 1817, but the assigned address is "<nobody@mess.org>", so I decided to post also here to bring more attention:

Genesis/Mega Drive driver regressed when MAME 0.128u7 was merged in SVN revision 3958. Maybe related with the bug of MAME megatech driver reported at http://www.mametesters.org/mantis/view.php?id=2817.

Some games only boot after the "9" key is pressed to reset them (e.g. Gaiares), various games need "9" to play sound (e.g. Puyo Puyo, Devil Crash, Raiden Trad, Sonic games with Knuckles, voices of Mortal Kombat 2 and Moonwalker), various have horrible sound, seems like playing wrong data (like Shadow Dancer when start 1st stage, Golden Axe 2 or JuJu Densetsu when an enemy dies - voice related?).

SVN rev. 3957 seems ok.
SVN rev. 3958 is the 0.128u7 merge.
SVN rev. 3963 made some changes to genesis.c related to Pico driver.
SVN rev. 3964 fixed the drivers requiring cartridges, when a regression in src/mess/image.c file was fixed, and the problems are first reproduced here. Later I tested without the (Pico) changes of the previous commit and there was no difference.

I take a look at the 0.128u7 commit. The MAME and MESS files related to the Genesis driver seems only adapted to the new core changes.

What seems strange to a non MESS/MAME developer like me are tags apparently unused, like the following part of src/mess/drivers/genesis.c:

ROM_START(genesis)
ROM_REGION(0x1415000, "maincpu", ROMREGION_ERASEFF)
ROM_REGION( 0x10000, "soundcpu", ROMREGION_ERASEFF)
ROM_END

The only reference to "soundcpu" (via a memory_region() call) I found was in src/mame/drivers/genesis.c, at READ8_HANDLER (genesis_z80_bank_r) which is only referenced at a disabled (with #if 0 .. #endif) address mapping in the same file.
Posted By: etabeta78

Re: SVN builds - new driver flood - 04/26/09 04:25 PM

I thought regression was due to the improvements to Yamaha sound core which had unexpected bad effects in MESS, for who knows what reason...
Posted By: R. Belmont

Re: SVN builds - new driver flood - 04/26/09 04:33 PM

The regression has nothing to do with the YM2612 improvements, all of which were tested in MESS originally and based directly on tests run on actual MD/Genesis hardware. The regression is purely a result of the u7 changes.

Anyway, you can "bring attention to" it all you want - the code is Haze's to fix and for hopefully obvious reasons nobody else wishes to touch it.

PS: "soundcpu" and "maincpu" are the names of the Z80 and 68000. There is an implicit mapping.
Posted By: judge

Re: SVN builds - new driver flood - 04/26/09 06:19 PM

At least in Zero Wing, it seems some weird z80 reset - 68k timing issue. The z80 seems to be running while the 68k is writing stuff to the shared ram but timing causes the z80 to screw up it's own ram in such a way that it enters an infinite loop.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 04/26/09 06:21 PM

Ahh. On real h/w the Z80 must be halted for the 68k to access that RAM. There's a register the 68k writes to halt the Z80 and grab it's bus. That must be what's gone wrong.
Posted By: judge

Re: SVN builds - new driver flood - 04/26/09 06:46 PM

Yeah, but inverting that behavior of that register does not fix it frown
Posted By: Haze

Re: SVN builds - new driver flood - 04/26/09 07:25 PM

Originally Posted By R. Belmont

Anyway, you can "bring attention to" it all you want - the code is Haze's to fix and for hopefully obvious reasons nobody else wishes to touch it.


I'm not in a position to be able to test and fix everything. I've documented the reasons for this before, ie, I have little desire to maintain trees for 2 projects, and can't work well without my fixed set lists.

For that reason, it's all yours / whoever wants to fix it. AFAIK it worked fine last time I checked it, and all the flags to halt / resume the z80 / 68k depending on accesses are full implemented. If somebody has broken it since then however, I can't tell you.
Posted By: etabeta78

Re: SVN builds - new driver flood - 04/26/09 11:24 PM

I remember your reasons, so I won't try to make you change idea. however, I wanted to add that, since last time you looked at it, I re-factored a bit the source and now MESS code is much closer to hazeMD.

MESS only defines the memory map and input handlers [1] in mess/drivers/genesis.c and then uses mame/drivers/megadriv.c for everything else. mess/machine/genesis.c only contains local code for cart loading and sram handling (which basically do not affect any other aspect of the main emulation).

therefore, if you ever get back to megadrive in MAME, MESS can inherit any change with no additional efforts!


[1] this is needed to offer options between 3-buttons pad and 6-buttons pad
Posted By: Haze

Re: SVN builds - new driver flood - 04/27/09 09:48 AM

I'm willing to bet that if I grab the current Mess code, and unzip it over my current MAME tree / any current SVN build, then all hell will break loose.

This means that unless I create a specific tree for working with Mess, which I don't sync to the MAME SVN, then I can't work on MESS. That means having to manually merge back in changes, and not being able to test the MAME/MESS side of things together in the same tree. It's also extra effort, and generally annoying.

The projects should co-exist in the same tree *somehow*, and should always compile, major changes should be made to both projects at the same time. Developing things which share code is more difficult than it needs to be otherwise.

For systems where it's appropriate MESS should have set lists like MAME in addition to the current open ended approach (which is ueful for homebrew & systems where the media isn't really 'fixed')

I don't dislike MESS (some of the recent changes have been very impressive thanks to Judge etc.) but it simply doesn't fit in my development methodology.

Sorry if others disagree, but that's just how I see things.

Posted By: etabeta78

Re: SVN builds - new driver flood - 04/27/09 10:21 AM

Originally Posted By Haze
I don't dislike MESS (some of the recent changes have been very impressive thanks to Judge etc.) but it simply doesn't fit in my development methodology.

Sorry if others disagree, but that's just how I see things.



This is perfectly fair. And as I said, I didn't want to force you to change your mind.

Originally Posted By Haze
I'm willing to bet that if I grab the current Mess code, and unzip it over my current MAME tree / any current SVN build, then all hell will break loose.


If you means: grab current MESS genesis code, you're true. you need a complete MESS tree to test the changes.

if you means: grab the whole MESS tree, you're wrong most of the times (unfortunately, these days you would be right due to the malloc changes)

my point was: if you keep two separate source folders, you make changes to megadriv.c in MAME and you copy it in the MAME part of MESS source, you can compile and test the results without the need of any change in the MESS part of the source (except if you want to change the memory map). earlier, you would have needed to tweak more MESS files.

In general, MESS source is sync'ed right after an intermediate update happens to MAME. this means you have de-sync'ed sources for more or less 7 days and it only greatly affects development when Aaron does huge changes (like the malloc ones). Usually, copying any MAME drivers and/or machine change (like any megadriv.c one) directly into the MAME part of MESS source and recompiling work like a charm.

Originally Posted By Haze

For systems where it's appropriate MESS should have set lists like MAME in addition to the current open ended approach (which is ueful for homebrew & systems where the media isn't really 'fixed')


As a first step in this direction, we could add a warning (a red screen?) when a known bad dump is loaded. We have .hsi files for reasons like these. And adding fixed lists to the current approach (without replacing it completely) is possible as well, it just requires some more coding.
Posted By: Haze

Re: SVN builds - new driver flood - 04/27/09 10:57 AM

Originally Posted By etabeta78

if you means: grab the whole MESS tree, you're wrong most of the times (unfortunately, these days you would be right due to the malloc changes)


replace 'these days' with 'quite often' and you'd be right. Significant changes are seem to be being made more often than not. As soon as I dump MESS in a MAME tree it means I can no longer update that MAME tree from SVN due to the chances of everything breaking.

This has been the case the last 5 or 6 times I've decided to take another look at MESS. Call me unlucky with my timing, but that's what I've found.

Posted By: judge

Re: SVN builds - new driver flood - 04/27/09 12:10 PM

There are also a couple of changes in mame files which are local to the mess tree. The number of changes is low and should eventually be cleaned up such that the differences are only in the mess files. One of these things is the 8042 keyboard controller code. These are usually not the cause of builds breaking though.

The bigger mame changes do tend to break building of mess but these issues usually get resolved by the time the regular uX update gets applied or reasonably quickly after the update is applied. From past experience, I'd say about 1 out of 3 uX updates break a straight mess build. So I can imagine Haze being unlucky with the updates of his source tree. Also there the situation is improving with cleaner code in mess and more shared devs between the two teams.

Anyway, back to genesis/megadrive: There are no obvious changes to the code, the reset handling all looks ok, so I don't have an idea what is wrong.

In Zero Wing I see the 68k switching off the z80, copying a bit of z80 code, turning the z80 on. When the 68k starts copying some more z80 code it seems the z80 _is_ running, but looking at the writes to the z80 reset register it should be running. Perhaps a bug in the original code which just happens to work on real systems?
Posted By: Haze

Re: SVN builds - new driver flood - 04/27/09 02:24 PM

a lot of the games have buggy code, yes. In some cases it just happened to work on the real hardware. (In others, it did if you were lucky, but if you ran the game on a different model it would fail)

One of the sound programs a lot of Accolade games use for example checks completely the wrong flag and instead of reading the z80 status, actually does an open bus read.

I wouldn't rule out Zero Wing having some buggy code, but OTOH I can't remember any problems with it.
Posted By: judge

Re: SVN builds - new driver flood - 04/27/09 04:20 PM

What Zero Wing seems to do is:
Code:
1. request z80 bus
2. clear z80 reset
3. copy inital z80 "ram clear program" to the z80 ram area
4. start z80 reset
4. return z80 bus
5. clear z80 reset
6. do some stuff
7. start z80 reset
8. request z80 bus
9. clear z80 reset
10. copy sound program to the z80 ram area
11. start z80 request
12. return z80 bus
13. clear z80 reset


Followed by a long list of 'return z80 bus' and 'clear z80 reset' messages which shouldn't do much since the lines are already released.

At point 3 however, the z80 is actually running. I'd say that that is incorrect, it should be halted/wiating for the bus to be released.


Posted By: R. Belmont

Re: SVN builds - new driver flood - 04/27/09 06:33 PM

Yeah, clearing reset AFAIK doesn't actually run the Z80 if the bus is held, so it shouldn't start running until step 5.
Posted By: Haze

Re: SVN builds - new driver flood - 04/27/09 06:54 PM

Originally Posted By R. Belmont
Yeah, clearing reset AFAIK doesn't actually run the Z80 if the bus is held, so it shouldn't start running until step 5.


From what I remember that's correct, it shouldn't run unless it's got the bus, and the reset line has been cleared.

In both reset high and non-bus state the z80 goes nowhere, and afaik that's how it used to behave.
Posted By: etabeta78

Re: SVN builds - new driver flood - 04/27/09 09:31 PM

speaking of broken systems, I tried to fix jaguar today (bug 1766). I came to the conclusion that somehow jaguar bios is loaded with the wrong endianness: if you start any jaguar cart, you see immediately that the CPU reads an illegal instruction rather than jumping to E00000 as expected.

culprit could be either ROM_LOAD (cojag uses a REGION32_BE) or the part of MACHINE_RESET(jaguar) where the first 0x10 bytes of the bios are copied to the shared RAM.

However, I haven't managed to fix it (probably because I used the wrong combination of ROM_REGION/ROM_LOAD....)
Has anyone any suggestion about this?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 04/27/09 10:39 PM

cojag uses a 32-bit processor. the endianness rules are different then.
Posted By: enik

Re: SVN builds - new driver flood - 04/28/09 01:52 AM

Originally Posted By judge
At least in Zero Wing, it seems some weird z80 reset - 68k timing issue.


I added a comment to bug 1817 which includes some differences when running on PAL or a NTSC driver.
Posted By: etabeta78

Re: SVN builds - new driver flood - 04/28/09 05:14 AM

Originally Posted By R. Belmont
cojag uses a 32-bit processor. the endianness rules are different then.


You mean that also the M68k variant which runs area51t and a few other cojag games is completely different from the jaguar?

To me, it still sounds like the jaguar in MESS hasn't been updated correctly at some point... looking at the debug, it seems the game code is now shifted back of 8 bits. I will dig into it a bit more, tonight
Posted By: R. Belmont

Re: SVN builds - new driver flood - 04/28/09 05:24 AM

eta: in MAME 16 and 32 bit processors store memory differently, even if they're the same family and endianness in real life. It's (part of) why it's harder than it should be to make a lot of things work on big-endian hosts.
Posted By: judge

Re: SVN builds - new driver flood - 04/28/09 05:50 AM

Originally Posted By Haze
Originally Posted By R. Belmont
Yeah, clearing reset AFAIK doesn't actually run the Z80 if the bus is held, so it shouldn't start running until step 5.


From what I remember that's correct, it shouldn't run unless it's got the bus, and the reset line has been cleared.

In both reset high and non-bus state the z80 goes nowhere, and afaik that's how it used to behave.


I have sent a patch to fix this to the mamedev mailinglist, awaiting comments...
Posted By: Justin

Re: SVN builds - new driver flood - 05/03/09 06:12 AM

Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/03/09 06:19 AM

Ahh, I knew Haze and Angelo's fixes would eventually get us someplace on that smile Fortunately the GD-ROM drive is ATAPI instead of the weird customness of the Saturn's so it should be much easier to get games running.
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/03/09 06:38 AM

speaking of saturn, do you have any idea of when the starting logo got corrupted?

and is the system loading CHDs in current svn? I don't have any file to test further the driver, but I was curious if regression only involved the logo or also the games
Posted By: mahlemiut

Re: SVN builds - new driver flood - 05/03/09 01:39 PM

Originally Posted By etabeta78
speaking of saturn, do you have any idea of when the starting logo got corrupted?

and is the system loading CHDs in current svn? I don't have any file to test further the driver, but I was curious if regression only involved the logo or also the games

The video issue has been there for a few versions now, from memory. As for CHDs, I don't know if the few I have are valid or not (since MESS is the only emulator that uses them, as far as I am aware), since they have never worked other than the QSound demo Arbee mentioned some time ago. From what I understand, the CD block needs a lot of work.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/03/09 01:51 PM

Now that we've started getting decaps back I'm just waiting on the SH-1 so we can have real correct CD block emulation smile

Bad graphics means it needs to get synced with the MAME stv.c again. (ETA: Just did so and it fixed it).
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/04/09 03:07 AM

Justin: how did you get that to work? On mine something goes wrong fairly quickly that the BIOS doesn't like and it crashes itself.
Posted By: Justin

Re: SVN builds - new driver flood - 05/04/09 04:13 AM

That screenshot is from dcdev, the other sets don't seem to get anywhere.
Posted By: incog

Re: SVN builds - new driver flood - 05/04/09 10:31 PM


Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/05/09 03:06 PM

Yeah, the AGA support is definitely not complete. Would be nice to start getting some of the demos that require an A1200 going though smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/06/09 06:24 PM

Be careful what you wish for or... you just may get it wink



Jokes aside, commit is in progress adding Amiga 1200 and CD32, both marked as Not Working. Thanks incog for the initial submission (adapted from cubocd32 in MAME)

I also hooked up the floppy drive to a1200, under the assumption it worked like on a a500... and I think I was wrong, because it doesn't load WorkBench.
In fact, reason may be anything: broken .adf image, mistakes in the CIA interface, or the floppy can require being hooked up to something completely different...

I'm not going to further work on this, because I have already too many things in progress (and I only wanted to reduce redundant code with MAME + add a1200), if anyone wants to step up and take care of this, the code is in the svn repository
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/07/09 07:43 AM

Thank you, to start with the A1200/CD32 driver support.

Posted By: Duke

Re: SVN builds - new driver flood - 05/07/09 01:42 PM

Well, you created a nice bastard there etabeta, a machine with an 68020 and AGA, using the 68000 kickstart from an A500 wink

The real A1200 used the following ROMs:

Kickstart 3.0: 391523-01.u6a and 391524-01.u6b
Kickstart 3.1: 391773-01.u6a and 391774-01.u6b
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/07/09 02:03 PM

that would explain a lot of things... as I said I just hooked up the basics (since we were adding cd32), but I haven't really investigated what a real a1200 would have needed.

would you mind to fix the kickstart?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/07/09 03:19 PM

You make me crazy. cry

I am just busy to find the daily needed new ...., no time for tests. grin
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/07/09 03:35 PM

Originally Posted By Duke
Well, you created a nice bastard there etabeta, a machine with an 68020 and AGA, using the 68000 kickstart from an A500 wink

The real A1200 used the following ROMs:

Kickstart 3.0: 391523-01.u6a and 391524-01.u6b
Kickstart 3.1: 391773-01.u6a and 391774-01.u6b


Hello Duke,

Kickstart 3.1: 391773-01.u6a = 40.068 ROM 0
Kickstart 3.1: 391774-01.u6b = 40.068 ROM 1
Kickstart 3.0: 391523-01.u6a = 39.106 ROM 0
Kickstart 3.0: 391524-01.u6b = 39.106 ROM 1

Is this correct ?

Posted By: Duke

Re: SVN builds - new driver flood - 05/07/09 03:45 PM

Anna: Yes, I think thats correct.

I'm not sure if I have the time to add them, maybe tomorrow.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/07/09 03:50 PM

Originally Posted By Duke
Anna: Yes, I think thats correct.

I'm not sure if I have the time to add them, maybe tomorrow.


Thank you. smile

Posted By: etabeta78

Re: SVN builds - new driver flood - 05/07/09 04:07 PM

Originally Posted By Duke
I'm not sure if I have the time to add them, maybe tomorrow.


if I have time I'll do it tonight. but feel free to fix any other mistake you may find
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/12/09 01:38 PM

Note to Lord Nightmare: all of the known-to-exist IIgs "Mark Twain" prototypes (the rumored "ROM 04" machines) had standard ROM 03 ROMs (yes someone dumped it and did a binary compare). The differences were purely hardware: more RAM and an on-board SCSI interface.
Posted By: incog

Re: SVN builds - new driver flood - 05/12/09 01:45 PM

A skeleton driver army just arrived in MESS SVN courtesy of Micko.

Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/12/09 09:05 PM

Originally Posted By R. Belmont
Note to Lord Nightmare: all of the known-to-exist IIgs "Mark Twain" prototypes (the rumored "ROM 04" machines) had standard ROM 03 ROMs (yes someone dumped it and did a binary compare). The differences were purely hardware: more RAM and an on-board SCSI interface.

Ok, i'll fix that comment/label when i convert 2gs to use the bios system. the number comes from a list of apple2 roms including several prototypes which are not dumped. i have no idea what that rom is supposed to be then.

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/12/09 09:45 PM

Please do not do a BIOS conversion on the IIgs. The hardware is actually different for each ROM version, although we don't emulate all the differences very well yet. (0/1 are mostly interchangable but ROM 03 and the "Mark Twain" are quite different).
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/13/09 12:17 AM

Ok. Should I do a bios select for just rom0/1 then? Or should I leave the driver alone? Also, do you have more info about which IIgs models used which keyboard MCU?

Lastly, what about bios system for IIe or other appleII variants?

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/13/09 01:33 AM

The IIe has no signficant hardware changes through it's lifetime so it can be BIOSed. The IIc did have hardware differences with the BIOS differences, which is why some models don't boot right now.

As far as the ADB MCU, http://www.llx.com/~nparker/a2/adb.html is *the* info source.
Posted By: mangamuscle

Re: SVN builds - new driver flood - 05/13/09 02:23 AM

I am not apple IIe expert (I was a commodore guy) but:

"In March 1985, Apple replaced the original machine with a new revision called the Enhanced IIe. This change involved a new processor, the CMOS based 65C02 CPU, a new character ROM for the text modes, and two new ROM firmware chips."

http://en.wikipedia.org/wiki/Apple_iie#The_Enhanced_IIe
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/13/09 02:55 AM

Right, but that was a software/firmware change. The hardware was the same aside from the CPU.
Posted By: mangamuscle

Re: SVN builds - new driver flood - 05/13/09 03:24 AM

That is my whole point, the basic difference betwen the 65c02 and the 6502 was to remove all the illegal opcodes (since otherwise there was no speed improvement). I have no idea if the pre-65c02 roms had any illegal opcodes in them, but one can be 100% sure without some testing, at least some game software did used the illegal opcodes in the 6502 and stopped working with the new cpu/firmware.
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/13/09 03:46 AM

R.Belmont: on that page (http://www.llx.com/~nparker/a2/adb.html), theres instructions to dump the mcu rom for the IIgs rom0/1 and IIgs rom3 machines (the roms are different on those two). Do you have access to a IIgs machine to do that to to obtain one of the roms? also, which of the two mcus is apple part number 341-0024-a?

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/13/09 04:13 AM

manga: the 'C02 also added some new instructions. ProDOS versions 2.0 and later used them to save memory and gain some performance.

LN: I have a working ROM 3 but no way to get data off of it other than by sight. (The 800k floppy drive works but I don't have a Mac of the required vintage to be able to convert that to a PC floppy right now - maybe I'll grab a cheap Classic off eBay or something).

The nice thing about that MCU is that it's the 8-bit parent of the M377xx series I emulated in MAME (except the 377xx added a second accumulator, which is kind of heresy but also useful).
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/13/09 06:06 AM

just to add my two cents: I think apple2c0, apple2c3, apple2c4 can be merged to use BIOS (but not with apple2c, due to the lack of IWM), the others may be left separate since they come from different machines, despite analogous but not identical hw (e.g. they have small input differences which would be lost)
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/13/09 06:11 AM

Originally Posted By R. Belmont
...

LN: I have a working ROM 3 but no way to get data off of it other than by sight. (The 800k floppy drive works but I don't have a Mac of the required vintage to be able to convert that to a PC floppy right now - maybe I'll grab a cheap Classic off eBay or something).

Why not send me the disk once you've dumped it? I have a catweasel which can read 800k format easily on a pc.

LN
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 05/13/09 10:34 AM

@ArBee: you can use a Sony USB floppy drive with a modern Mac to read 800k floppies.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/13/09 02:40 PM

You sure? Everything I see online says USB drives only can read Mac or PC 1.44MB and PC 720k disks, not the variable speed 800k (or 400k) ones.
Posted By: mangamuscle

Re: SVN builds - new driver flood - 05/13/09 09:53 PM

If someone can confirm such USB floppy drives exist would it be possible to post exact brand and model number?
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 05/14/09 09:04 AM

My friend bought one a few years ago (about 2003 or 2004, I think) - it was Sony-branded and advertised 800kB compatibility as a feature.
Posted By: Heihachi_73

Re: SVN builds - new driver flood - 05/14/09 03:50 PM

This might sound like the El Paso flat bottomed taco ad, but...

Can't you just use an old Apple or classic Mac to dump 800k disks to an image format, and copy the image across to a later, somewhat usable machine using a network/serial cable? I gather there has to be at least some sort of network connection for them, even if it does take 30 minutes to copy a file across!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/14/09 04:05 PM

Sure, you can do that if you have the right cables and such. A classic Mac with a SuperDrive can just read the 800k disks and write the image to PC-formatted 1.44MB disks though, which would be probably less hassle smile
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/14/09 04:33 PM

Yeah, but you lose the sector accuracy unless you copy it as a .image file or something.

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/14/09 05:08 PM

I thought this was about getting the ADB microcontroller BIOS off my IIgs, not sector-accurate anything smile
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/17/09 03:24 AM

In this case, this is true. that message was for doing apple dumps in general. (also you lose the file info-tree if you copy it from an hfs filesystem to a fat filesystem *I THINK*, unless you store the metadata in a separate file)

LN
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/17/09 04:24 PM

It exist massive skeleton driver updates for the last SVN builds. I know, they are not working yet.
My question:
if i get error if i try to load a skeleton driver, make it sense to post this error or not ?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/17/09 04:28 PM

Skeletons generally shouldn't crash, that's about the only rule with them.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/17/09 06:09 PM

SVN revision 4723

Windows GUI

If I try to start any FM-Towns skeleton driver I get following error message:

i386: Invalide opcode FF at FFFFFFF1

Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/17/09 08:27 PM

I test more than 60 new skeleton driver. It seems, I have only the problem with the FM-Towns/Marty skeleton driver.
Posted By: MooglyGuy

Re: SVN builds - new driver flood - 05/17/09 10:39 PM

That isn't a crash, it's a fatal error.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/17/09 10:49 PM

I don't know about MESSUI, but in commandline versions a fatal error may as well be a crash.
Posted By: MooglyGuy

Re: SVN builds - new driver flood - 05/17/09 10:59 PM

I don't know about you, but given that the only alternative is to be completely silent and/or to actually segfault, I would much rather the i386 core call fatalerror() when it hits an unknown opcode due to a driver's skeletal nature causing the BIOS run off into the weeds.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/18/09 06:45 AM

I tried to start the driver on commandline too.

i386: Invalide opcode FF at 000FC001



New SVN build 4732 (Pentium default/Pentium with symbols) from Bobz.
My test computer:
DELL C840
Mobile Intel(R) Pentium(R) 4 - M CPU 1.80GHz
WinXP Pro SP2



Posted By: etabeta78

Re: SVN builds - new driver flood - 05/18/09 07:09 AM

yeah, it's a problem of incomplete i386 emulation. as long as there are no improvements into the MAME core, the FM Towns driver will keep crashing.

it is one of the reasons this is only a skeleton wink
Thanks for testing

Posted By: mahlemiut

Re: SVN builds - new driver flood - 05/19/09 12:32 PM

The fatal error occurs whenever the program starts running on some part of RAM or ROM that it shouldn't. Anyway, it now runs a touch longer (about one second ;)) before a fatal error.

Hopefully, protected mode isn't needed for booting (and it has MS-DOS on ROM, so shouldn't be needed there either), so maybe one day it'll do something useful. smile Not that figuring out all the hardware that it does have and how it works won't be easy.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/19/09 02:22 PM

Originally Posted By mahlemiut
The fatal error occurs whenever the program starts running on some part of RAM or ROM that it shouldn't. Anyway, it now runs a touch longer (about one second ;)) before a fatal error.
Hopefully, protected mode isn't needed for booting (and it has MS-DOS on ROM, so shouldn't be needed there either), so maybe one day it'll do something useful. smile Not that figuring out all the hardware that it does have and how it works won't be easy.


Great, you make me for one second happy. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/19/09 03:01 PM

Thanks for the love for the Japan-only systems, Barry. They're great but the Japanese emulators are often incomprehensible. Being able to play the X68000 games in MESS is much handier, and it'll be nice to have FM Towns going eventually too smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/19/09 03:48 PM

the main problem is finding reliable BIOS dumps (actually, even unreliable ones are hard find...)

has anyone ever tried to contact Team Japump!!! ? maybe they could keep their eyes open...
Posted By: Darkstar

Re: SVN builds - new driver flood - 05/19/09 08:12 PM

I will be in Japan from October to February, and I could try to look for some missing systems (from the miss list in the wiki) there... I can't promise anything though

--Darkstar
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 05/19/09 11:59 PM

Originally Posted By Darkstar
I will be in Japan from October to February, and I could try to look for some missing systems (from the miss list in the wiki) there... I can't promise anything though

--Darkstar


A lot of the missing ones are probably pretty obscure, but I'd love to see some of these: Bandai Supervision 8000 (1979) and games, more games for Gakken TV Boy Compact Vision (1983), My Vision (Nichibutsu, 1983) and games, Casio PV-1000 (1983) and games, and missing stuff for Epoch Cassette Vision and Super Cassette Vision. That's a quick list, so I could come up with some more details if you have questions.
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/20/09 12:41 AM

Also any of the epson HC-xx, PC-xx or QC-xx systems.
(released in europe and us as HX-xx (i.e. hx-20) PX-xx and QX-xx (i.e. QX-10))

LN
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/20/09 10:37 AM

Originally Posted By ranger_lennier
Originally Posted By Darkstar
I will be in Japan from October to February, and I could try to look for some missing systems (from the miss list in the wiki) there... I can't promise anything though

--Darkstar


A lot of the missing ones are probably pretty obscure, but I'd love to see some of these: Bandai Supervision 8000 (1979) and games, more games for Gakken TV Boy Compact Vision (1983), My Vision (Nichibutsu, 1983) and games, Casio PV-1000 (1983) and games, and missing stuff for Epoch Cassette Vision and Super Cassette Vision. That's a quick list, so I could come up with some more details if you have questions.


I agree. For Epoch Cassette Vision as example, I found only one game (Wheelie Racer [d1].scv) in internet, but no dumped bios. Maybe someone can ask Takeda Toshiya, not sure he will help ?
Posted By: Darkstar

Re: SVN builds - new driver flood - 05/20/09 02:41 PM

OK, I noted that down and will look for those.

I'm personally interested in getting an FM-Towns and maybe an X68000 as well (for my own collection), are there any specific models of those that I should keep an eye out for? Anything that's needed? (although I doubt that they will be available in such an abundance that I can cherry-pick what I want... I'll probably take what I can get ;-)
Posted By: mahlemiut

Re: SVN builds - new driver flood - 05/20/09 09:44 PM

An X68000 Super or an X68000 XVI (or later) would be useful, as these have a built-in SCSI controller that we need a ROM dump from (should be possible in software, just got to transfer it somehow once read)
Posted By: Shideravan

Re: SVN builds - new driver flood - 05/21/09 01:14 AM

I increment with the following systems:

-PV2000 (computer)
-Bit Corp. Gamate
-Tomy Kiss site
-Bandai Telebiko
-Nintendo Super Famicom Satellaview

And, of course, i'm assuming that this part of the wiki is updated :p

http://mess.redump.net/dumping:wanted?s[]=pocket&s[]=computer
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 05/21/09 01:29 AM

Originally Posted By Anna Wu

I agree. For Epoch Cassette Vision as example, I found only one game (Wheelie Racer [d1].scv) in internet, but no dumped bios. Maybe someone can ask Takeda Toshiya, not sure he will help ?


Probably not coincidence that that's the cart I sent Guru. smile I sent an email to Takeda Toshiya once, but no response. Maybe someone else could try, especially someone in Japan. We are looking at duplicating a lot of work that's already been done collecting and dumping games, because the dumps just don't seem to be available.
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 05/21/09 01:31 AM

Originally Posted By Shideravan
I increment with the following systems:

-PV2000 (computer)
-Bit Corp. Gamate
-Tomy Kiss site
-Bandai Telebiko
-Nintendo Super Famicom Satellaview

And, of course, i'm assuming that this part of the wiki is updated :p

http://mess.redump.net/dumping:wanted?s[]=pocket&s[]=computer


The Gamate's European, not Japanese. I have some stuff for it, but it seems the carts are quite difficult to dump.
Posted By: incog

Re: SVN builds - new driver flood - 05/21/09 02:13 AM

Also the Nintendo Super Famicom Satellaview is dumped and the Bandai Telebiko isnt what you think it is, it's a kids toy, the lightgun vhs-tape thing it's labeled as on a few sites is wrong. That's actually a Bandai Video Challenger. I have 2 of those crappy things, both of mine are the European version.
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/21/09 06:06 AM

my gamate is going to reach a dumper's hands soon... hopefully it can be dumped
Posted By: Anna Wu

Re: SVN builds - new driver flood - 05/21/09 06:22 AM

Quote:
Probably not coincidence that that's the cart I sent Guru.


Can be. smile

Quote:
I sent an email to Takeda Toshiya once, but no response.


It seems, he keeps his ROM´s like a Holy Grail.
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/21/09 07:25 AM

no. the point is that Japanese people take very seriously the legal issues related to dumping. no matter how old a piece of hardware is.

that's why you seldom find dumps from Japan released by Japanese people.
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 05/21/09 12:56 PM

Originally Posted By etabeta78
my gamate is going to reach a dumper's hands soon... hopefully it can be dumped


Nice! If he has any luck, I can send my carts if any are needed.
Posted By: Shideravan

Re: SVN builds - new driver flood - 05/21/09 03:08 PM

Originally Posted By incog
Also the Nintendo Super Famicom Satellaview is dumped and the Bandai Telebiko isnt what you think it is, it's a kids toy, the lightgun vhs-tape thing it's labeled as on a few sites is wrong. That's actually a Bandai Video Challenger. I have 2 of those crappy things, both of mine are the European version.


I ever thought that Telebiko was a japanese version of V. Challenger...

Living and learning...
Posted By: Justin

Re: SVN builds - new driver flood - 05/22/09 01:11 AM

Looks from that Telebiko picture like it displays on the TV, so it's probably emulatable unless it's a horrible VHS bodge...may not be terribly fun unless you're a Japanese kid though.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 05/23/09 01:31 PM

Originally Posted By Darkstar
I will be in Japan from October to February, and I could try to look for some missing systems (from the miss list in the wiki) there... I can't promise anything though


It would be nice to get these models with Japanese symbols on the keyboard:

NEC PC-8201
NEC PC-8300
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/23/09 10:50 PM

cybiko drivers have been enabled again in MESS. Notice that cybikoxt crashes almost immediately due to unemulated opcodes. This is the expected behavior until H8S2xxxx support will be added to the H83002 core.

cybikov1 & v2 would crash as well for the same reason if I would have not unmapped the ROM from the address maps (it can be easily re-added once the CPU is supported)
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/23/09 11:18 PM

I don't know if it's a great idea to enable drivers we don't have a CPU core for - it's just gonna crash and make regression testing impossible.
Posted By: Robbbert

Re: SVN builds - new driver flood - 05/23/09 11:49 PM

I tend to agree... or at least do something so it doesn't crash.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/24/09 01:43 AM

BTW, I'd love to work on adding H8S to the H8 core right now but SVN is busted.

Quote:
svn: Can't find a temporary directory: Internal error


...which Google says means the server is out of disk space.

This is a limited time offer before I have to go back to ignoring MESS, so hopefully it gets fixed soon ;-)
Posted By: Robbbert

Re: SVN builds - new driver flood - 05/24/09 03:01 AM

You can PM or send to me if you like.. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/24/09 06:39 AM

Originally Posted By R. Belmont
I don't know if it's a great idea to enable drivers we don't have a CPU core for - it's just gonna crash and make regression testing impossible.


well, cybiko v1/v2 run for at least 10 seconds without problems (but of course they're not executing any code, so they only show a blank screen); cybikoxt acts like fmtowns was acting a few days ago: it starts, it runs a couple of seconds, and finally it gets an exception due to unimplemented opcode

as long as we ignore its output error, regression testing should not be harmed by its presence wink
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/24/09 06:47 AM

Well, like I said, I'm happy to fix that if someone gets SVN working again =) I just did a lot of H8 work for System 23 in MAME (you'll see it in the next u release).
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/24/09 07:03 AM

just in case svn does not come back soon, you can use the latest revision you have: the driver was always compiled and only the messdriv.c entries were commented out. uncomment them and you can try those systems with any updated CPU core you may have

(notice that the only changes I have done to enable cybiko are the following
http://git.redump.net/cgit.cgi/mess/commit/?id=eda30fc4edf9af9e90de5eea9452dd9c911aabe9
a couple of defines for CPU_H8Sxxx = CPU_H83002, and a couple of #if 0 in address maps... I suggest you to keep only the tag fix flash1->flash2 of all those changes)
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 05/24/09 04:44 PM

Originally Posted By R. Belmont
Well, like I said, I'm happy to fix that if someone gets SVN working again =) I just did a lot of H8 work for System 23 in MAME (you'll see it in the next u release).


SVN back. Apparently server did not have disk full but svn server process was crashed/hung.

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/24/09 04:45 PM

Cool, thanks LN. It's indeed working smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/24/09 08:18 PM

It's not actually missing opcodes, it's that it puts the stack on internal RAM that's in a different place from the 3002 so an RTS goes berzerk very soon smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/31/09 09:42 PM

Well, well. (Not the dev BIOS either - that's the real US one).



In fairness, co-credit goes to Deunan Knute, Haze, Angelo Salese, and Olivier Galibert.
Posted By: etabeta78

Re: SVN builds - new driver flood - 05/31/09 09:49 PM

w00t!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 05/31/09 09:58 PM

And a shoutout to our PAL homies.

Posted By: Stiletto

Re: SVN builds - new driver flood - 06/02/09 01:47 AM

sweet!
Posted By: Just Desserts

Re: SVN builds - new driver flood - 06/02/09 02:52 AM

If anyone wants to try it out, I just committed my CPU core and driver for the Manchester Small-Scale Experimental Machine (SSEM), or "Baby", to the MESS SVN depot. It currently runs all known SSEM programs bundled with David Sharp's SSEM simulator, available here.

I am not entirely happy with the fact that it is compatible with all of the programs, though. Certain programs in particular, i.e. "nightmare.snp", would not run on the SSEM had it ever been extended to the full 8192 words of storage space of which it was theoretically capable (per some SSEM history sites), as they pad out the unused 8 address bits with pretty patterns.

Edit: Pretty picture:



The controls are as follows:
Up / Down: Move the selected store line up/down
Button 1: Halt / un-halt the SSEM
1-8, Q-I, A-K, Z-,: Toggle bits 0-31 of the currently-selected store line
Posted By: Stiletto

Re: SVN builds - new driver flood - 06/03/09 01:39 PM

Excellent work!

Maybe the Mailüfterl should be next!

Of course I don't know anything about it, but Norbert Kehrer emulated/simulated it.
http://web.utanet.at/nkehrer/mailuefterl/index.html
Posted By: Just Desserts

Re: SVN builds - new driver flood - 06/04/09 01:43 AM

Actually, I just had a browse through BitSavers, and apparently there's full ROM dumps, full hardware documentation, and some software dumps for the Xerox Alto. Might be fun to try to emulate the machine that indirectly gave rise to the Macintosh as we know it.
Posted By: Stiletto

Re: SVN builds - new driver flood - 06/04/09 12:37 PM

While I'm sure you're doing the research, I thought I'd supply the following links:
http://altogether.brouhaha.com/
http://toastytech.com/guis/gv.html
http://woodward.org/mps/

All kinda interesting. Also looks like you might want to talk to Juergen. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 06/04/09 12:49 PM

talking about Juergen and Alto, let me add these links then

http://web.archive.org/web/20070623191927/pullmoll.stop1984.com/alto/

http://web.archive.org/web/20070817034757/pullmoll.stop1984.com/alto/alto.txt
Posted By: incog

Re: SVN builds - new driver flood - 06/04/09 01:39 PM

The Alto was way ahead of its time.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/04/09 01:56 PM

If Juergen can't make it boot I don't want anything to do with it. That's a 5-Aarons-with-devil-horns difficulty level smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 06/04/09 04:00 PM

Originally Posted By R. Belmont
If Juergen can't make it boot I don't want anything to do with it. That's a 5-Aarons-with-devil-horns difficulty level smile


Dunno where you heard that he couldn't make it boot, according to his Alto dev blog - which is apparently only up when he's online, hence why it wouldn't show up on archive.org - says he got it up and running well enough to boot run some Alto games, and even for the ethernet card to receive packets. I'll see if I can't mirror it on my webspace.

Edit: http://moogle-tech.com/juergen/alto/
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/05/09 11:07 AM

I like the good progress for the FM-7 emulation. smile
Cannot wait for the actual SVN version to see the right color palette on Thunder Force game.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/06/09 07:42 AM

SVN r4949
FM-7

The color palette for the FM-7 emulation looks good.

1st picture: XM7 V3.4 L22
2nd picture: MESS
Game: Thunder Force (Tape)


Posted By: etabeta78

Re: SVN builds - new driver flood - 06/06/09 10:13 AM

yeah, astounding progresses indeed!! I think we have the first emulation ever of the FM7 on Linux & MacOSX smile

many thanks to mahlemiut: I was quite curious about this system and now I can finally try it!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/07/09 01:03 AM

Yes, this is MESS WIP, not MAME. (SVN #4957 or better, please).


Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/07/09 01:10 AM

For NAM-1975, paste the following into a file named layout.xml. Then add it to a copy of your MAME 0.131u4 approved nam1975.zip, rename the .zip as nam1975.rpk, and mess aes -cart nam1975.rpk.

Code:
<?xml version="1.0" encoding="utf-8"?>
<romset version="1.0">
   <resources>
      <rom id="program" file="001-p1.bin" crc="cc9fc951" sha1="92f4e6ddeeb825077d92dbb70b50afea985f15c0"/>
      <rom id="tiles" file="001-s1.bin" crc="7988ba51" sha1="bc2f661f381b06b34ac2fa215dd5689d3bf84832"/>
      <rom id="z80prg" file="001-m1.bin" crc="ba874463" sha1="a83514f4b20301f84a98699900e2593f1c1b8846"/>
      <rom id="ymsamples" file="001-v1.bin" crc="a7c3d5e5" sha1="e3efc86940f91c53b7724c4566cfc21ea1a7a465"/>
      <rom id="ymdelta1" file="001-v2.bin" crc="55e670b3" sha1="a047049646a90b6db2d1882264df9256aa5a85e5"/>
      <rom id="ymdelta2" file="001-v22.bin" crc="ab0d8368" sha1="404114db9f3295929080b87a5d0106b40da6223a"/>
      <rom id="ymdelta3" file="001-v23.bin" crc="df468e28" sha1="4e5d4a709a4737a87bba4083aeb788f657862f1a"/>
      <rom id="sprite1" file="001-c1.bin" crc="32ea98e1" sha1="c2fb3fb7dd14523a4b4b7fbdb81f44cb4cc48239"/>
      <rom id="sprite2" file="001-c2.bin" crc="cbc4064c" sha1="224c970fd060d841fd430c946ef609bb57b6d78c"/>
      <rom id="sprite3" file="001-c3.bin" crc="0151054c" sha1="f24fb501a7845f64833f4e5a461bcf9dc3262557"/>
      <rom id="sprite4" file="001-c4.bin" crc="0a32570d" sha1="f108446ec7844fde25f7a4ab454f76d384bf5e52"/>
      <rom id="sprite5" file="001-c5.bin" crc="90b74cc2" sha1="89898da36db259180e5261ed45eafc99ca13e504"/>
      <rom id="sprite6" file="001-c6.bin" crc="e62bed58" sha1="d05b2903b212a51ee131e52c761b714cb787683e"/>
   </resources>
   <configuration>
      <pcb type="standard">
         <socket id="p1" uses="program"/>
         <socket id="s1" uses="tiles"/>
         <socket id="m1" uses="z80prg"/>
	 <socket id="v11" uses="ymsamples"/>
	 <socket id="v21" uses="ymdelta1"/>
	 <socket id="v22" uses="ymdelta2"/>
	 <socket id="v23" uses="ymdelta3"/>
	 <socket id="c1" uses="sprite1"/>
	 <socket id="c2" uses="sprite2"/>
	 <socket id="c3" uses="sprite3"/>
	 <socket id="c4" uses="sprite4"/>
	 <socket id="c5" uses="sprite5"/>
	 <socket id="c6" uses="sprite6"/>
      </pcb>
   </configuration>
</romset>
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/07/09 01:16 AM

Samurai Shodown: take samsho.zip and samshoh.zip from MAME 0.131u4. Combine all the contents into samsho.zip. Add the layout.xml listed here then rename as .rpk once again and run in MESS.

Code:
<?xml version="1.0" encoding="utf-8"?>
<romset version="1.0">
   <resources>
      <rom id="program" file="045-p1.bin" crc="dfe51bf0" sha1="2243af3770a516ae698b69bcd9daf53632d9128d"/>
      <rom id="program2" file="045-p2.bin" crc="38ee9ba9" sha1="48190699a6be83cb6257365ae81f93fdd23abe09"/>
      <rom id="tiles" file="045-s1.bin" crc="9142a4d3" sha1="54088e99fcfd75fd0f94852890a56350066a05a3"/>
      <rom id="z80prg" file="045-m1.bin" crc="95170640" sha1="125c502db0693e8d11cef619b090081c14a9a300"/>
      <rom id="ymsamples" file="045-v1.bin" crc="37f78a9b" sha1="6279b497d12fa90b49ab5ac3aae20fb302ec8b81"/>
      <rom id="ymsamples2" file="045-v2.bin" crc="568b20cf" sha1="61af858685472a1fad608e230cccc2b108509ddb"/>
      <rom id="sprite1" file="045-c1.bin" crc="2e5873a4" sha1="65c74c1e2d34390666bbb630df7d1f4c9570c3db"/>
      <rom id="sprite2" file="045-c2.bin" crc="04febb10" sha1="16a8cbf0fd9468e81bf9eab6dbe7a8e3623a843e"/>
      <rom id="sprite3" file="045-c3.bin" crc="f3dabd1e" sha1="c80e52df42be9f8b2e89b467b11ab140a480cee8"/>
      <rom id="sprite4" file="045-c4.bin" crc="935c62f0" sha1="0053d40085fac14096b683f4341f65e543b71dc1"/>
      <rom id="sprite5" file="045-c5.bin" crc="a2bb8284" sha1="aa118e3b8c062daa219b36758b9a3814c08c69dc"/>
      <rom id="sprite6" file="045-c6.bin" crc="4fa71252" sha1="afe374a9d1f2d955a59efe7b6196b89e021b164c"/>
   </resources>
   <configuration>
      <pcb type="standard">
         <socket id="p1" uses="program"/>
         <socket id="p2" uses="program2"/>
         <socket id="s1" uses="tiles"/>
         <socket id="m1" uses="z80prg"/>
	 <socket id="v11" uses="ymsamples"/>
	 <socket id="v12" uses="ymsamples2"/>
	 <socket id="c1" uses="sprite1"/>
	 <socket id="c2" uses="sprite2"/>
	 <socket id="c3" uses="sprite3"/>
	 <socket id="c4" uses="sprite4"/>
	 <socket id="c5" uses="sprite5"/>
	 <socket id="c6" uses="sprite6"/>
      </pcb>
   </configuration>
</romset>
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/07/09 01:17 AM

Additional XMLs can be posted on the wiki or something, I would imagine. Note that not all games in MAME will work right now (anything with overly fancy ROM loading or encrypted ROMs is right out), and not all dumps in MAME are AES. Stick with ones marked "AES VERSION" or "MVS AND AES VERSION" in neodrvr.c.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/07/09 03:50 AM

Thanks Arbee, the 2 games are running well. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/07/09 04:20 AM

One more: tpgolf. Same procedure as the other two, and here's the layout.xml:

Code:
<?xml version="1.0" encoding="utf-8"?>
<romset version="1.0">
   <resources>
      <rom id="program" file="003-p1.bin" crc="f75549ba" sha1="3f7bdf5e2964e921fe1dd87c51a79a1a501fc73f"/>
      <rom id="program2" file="003-p2.bin" crc="b7809a8f" sha1="1604c889592c9610668bff296de48a0d6906156d"/>
      <rom id="tiles" file="003-s1.bin" crc="7b3eb9b1" sha1="39cd8bad9f8bfdeb8ac681b5b79ae5aa81c8dd5f"/>
      <rom id="z80prg" file="003-m1.bin" crc="4cc545e6" sha1="8e014b8cab3e5b3995756a4ea52ce49c36866377"/>
      <rom id="ymsamples" file="003-v1.bin" crc="ff97f1cb" sha1="defa249d46ae220f7bfa70746f5202bbbcc3e5fe"/>
      <rom id="ymdelta1" file="003-v2.bin"  crc="d34960c6" sha1="36d5877d5e42aab943f4d693159f4f3ad8b0addc"/>
      <rom id="ymdelta2" file="003-v22.bin" crc="9a5f58d4" sha1="2b580595e1820430a36f06fd3e0e0b8f7d686889"/>
      <rom id="ymdelta3" file="003-v23.bin" crc="30f53e54" sha1="22461f88a56d272b78dbc23204c0c6816200532b"/>
      <rom id="ymdelta4" file="003-v24.bin" crc="5ba0f501" sha1="ca02937a611a2c50c9e4b54f8fd4eaea09259894"/>
      <rom id="sprite1" file="003-c1.bin" crc="0315fbaf" sha1="583c9253219c1026d81ee5e0cf5568683adc2633"/>
      <rom id="sprite2" file="003-c2.bin" crc="b4c15d59" sha1="b0d8ec967f9b8e5216301c10b2d36912abce6515"/>
      <rom id="sprite3" file="003-c3.bin" crc="8ce3e8da" sha1="bc6c49b27d498f75a0d1a8c4d0cca75e140b9efc"/>
      <rom id="sprite4" file="003-c4.bin" crc="29725969" sha1="f1407da84919c2b3fe0e8f1fca65934b147c86c7"/>
      <rom id="sprite5" file="003-c5.bin" crc="9a7146da" sha1="2fc83d13e3e9565919aab01bf2a1b028f433b547"/>
      <rom id="sprite6" file="003-c6.bin" crc="1e63411a" sha1="ee397e2f679042e87b37d95837af62bb95a72af9"/>
      <rom id="sprite7" file="003-c7.bin" crc="2886710c" sha1="1533dd935f0a8f92a0a3c47d1d2bc6d035454244"/>
      <rom id="sprite8" file="003-c8.bin" crc="422af22d" sha1="f67c844c34545de6ea187f5bfdf440dec8518532"/>
   </resources>
   <configuration>
      <pcb type="standard">
         <socket id="p1" uses="program"/>
         <socket id="p2" uses="program2"/>
         <socket id="s1" uses="tiles"/>
         <socket id="m1" uses="z80prg"/>
	 <socket id="v11" uses="ymsamples"/>
	 <socket id="v21" uses="ymdelta1"/>
	 <socket id="v22" uses="ymdelta2"/>
	 <socket id="v23" uses="ymdelta3"/>
	 <socket id="v24" uses="ymdelta4"/>
	 <socket id="c1" uses="sprite1"/>
	 <socket id="c2" uses="sprite2"/>
	 <socket id="c3" uses="sprite3"/>
	 <socket id="c4" uses="sprite4"/>
	 <socket id="c5" uses="sprite5"/>
	 <socket id="c6" uses="sprite6"/>
	 <socket id="c7" uses="sprite7"/>
	 <socket id="c8" uses="sprite8"/>
      </pcb>
   </configuration>
</romset>
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/07/09 04:38 AM

Just a screenshot. smile

Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/07/09 04:56 AM

Thanks Anna smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/07/09 05:57 AM

Ghost Pilots



My layout.xml :
Code:
<?xml version="1.0" encoding="utf-8"?>
<romset version="1.0">
   <resources>
      <rom id="program" file="020-p1.bin" crc="e6f2fe64" sha1="50ab82517e077727d97668a4df2b9b96d2e78ab6"/>
      <rom id="program2" file="020-p2.bin" crc="edcb22ac" sha1="505d2db38ae999b7d436e8f2ff56b81796d62b54"/>
      <rom id="tiles" file="020-s1.bin" crc="a6d83d53" sha1="9a8c092f89521cc0b27a385aa72e29cbaca926c5"/>
      <rom id="z80prg" file="020-m1.bin" crc="48409377" sha1="0e212d2c76856a90b2c2fdff675239525972ac43"/>
      <rom id="ymsamples" file="020-v11.bin" crc="1b526c8b" sha1="2801868d2badcf8aaf5d490e010e4049d81d7bc1"/>
      <rom id="ymdelta1" file="020-v12.bin" crc="4a9e6f03" sha1="d3ac11f333b03d8a318921bdaefb14598e289a14"/>
      <rom id="ymdelta2" file="020-v21.bin" crc="7abf113d" sha1="5b2a0e70f2eaf4638b44702dacd4cb17838fb1d5"/>
      <rom id="sprite1" file="020-c1.bin" crc="bd6fe78e" sha1="50b704862cd79d64fa488e621b079f6e413c33bc"/>
      <rom id="sprite2" file="020-c2.bin" crc="5f4a925c" sha1="71c5ef8141234daaa7025427a6c65e79766973a5"/>
      <rom id="sprite3" file="020-c3.bin" crc="d1e42fd0" sha1="f0d476aebbdc2ce008f5f0783be86d295b24aa44"/>
      <rom id="sprite4" file="020-c4.bin" crc="edde439b" sha1="79be7b10ecdab54c2f77062b8f5fda0e299fa982"/>
   </resources>
   <configuration>
      <pcb type="standard">
   <socket id="p1" uses="program"/>
   <socket id="p2" uses="program2"/>
   <socket id="s1" uses="tiles"/>
   <socket id="m1" uses="z80prg"/>
   <socket id="v11" uses="ymsamples"/>
   <socket id="v12" uses="ymdelta1"/>
   <socket id="v21" uses="ymdelta2"/>
   <socket id="c1" uses="sprite1"/>
   <socket id="c2" uses="sprite2"/>
   <socket id="c3" uses="sprite3"/>
   <socket id="c4" uses="sprite4"/>
      </pcb>
   </configuration>
</romset>
Posted By: Shideravan

Re: SVN builds - new driver flood - 06/07/09 11:23 AM

Good work, the cartridges that I tested is running smooth as MAME!
Any purpose to MESS itself includes this xmls?
Posted By: judge

Re: SVN builds - new driver flood - 06/07/09 12:10 PM

RB: nice work! Next is neogeo cd? whistle
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/07/09 02:24 PM

CD's a bit of a pain - as far as I can tell, all existing emulators operate by pulling certain commands and stuff out of RAM rather than actually emulating the hardware registers.
Posted By: judge

Re: SVN builds - new driver flood - 06/07/09 04:11 PM

sounds like a challenge laugh
Posted By: incog

Re: SVN builds - new driver flood - 06/07/09 04:35 PM

atleast the Cd controlled used by the neogeo is the same as the sega cd, two birds one stone
Posted By: mahlemiut

Re: SVN builds - new driver flood - 06/08/09 12:18 AM

The Super Spy
Code:
<?xml version="1.0" encoding="utf-8"?>
<romset version="1.0">
   <resources>
      <rom id="program" file="011-p1.bin" crc="c7f944b5" sha1="da7560e09187c68f1d9f7656218497b4464c56c9"/>
      <rom id="program2" file="011-sp2.bin" crc="811a4faf" sha1="8169dfaf79f52d80ecec402ce1b1ab9cafb7ebdd"/>
      <rom id="tiles" file="011-s1.bin" crc="ec5fdb96" sha1="8003028025ac7bf531e568add6ba66c02d0b7e84"/>
      <rom id="z80prg" file="011-m1.bin" crc="ca661f1b" sha1="4e3cb57db716ec48487c1b070c3a55a5faf40856"/>
      <rom id="ymsamples" file="011-v11.bin" crc="5c674d5c" sha1="d7b9beddeb247b584cea9ca6c43ec6869809b673"/>
      <rom id="ymsamples2" file="011-v12.bin" crc="9f513d5a" sha1="37b04962f0b8e2a74abd35c407337a6151dc4e95"/>
      <rom id="ymdelta1" file="011-v21.bin" crc="426cd040" sha1="b2b45189837c8287223c2b8bd4df9525b72a3f16"/>
      <rom id="sprite1" file="011-c1.bin" crc="cae7be57" sha1="43b35b349594535689c358d9f324adda55e5281a"/>
      <rom id="sprite2" file="011-c2.bin" crc="9e29d986" sha1="b417763bad1acf76116cd56f4203c2d2677e22e5"/>
      <rom id="sprite3" file="011-c3.bin" crc="14832ff2" sha1="1179792d773d97d5e45e7d8f009051d362d72e24"/>
      <rom id="sprite4" file="011-c4.bin" crc="b7f63162" sha1="077a81b2bb0a8f17c9df6945078608f74432877a"/>
   </resources>
   <configuration>
      <pcb type="standard">
         <socket id="p1" uses="program"/>
         <socket id="p2" uses="program2"/>
         <socket id="s1" uses="tiles"/>
         <socket id="m1" uses="z80prg"/>
	 <socket id="v11" uses="ymsamples"/>
	 <socket id="v12" uses="ymsamples2"/>
	 <socket id="v21" uses="ymdelta1"/>
 	 <socket id="c1" uses="sprite1"/>
	 <socket id="c2" uses="sprite2"/>
	 <socket id="c3" uses="sprite3"/>
	 <socket id="c4" uses="sprite4"/>
      </pcb>
   </configuration>
</romset>



I've also tried Fatal Fury 3, and I get a segfault during initialisation.
Code:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f9e273e86f0 (LWP 7247)]
0x0000000000db8f11 in astring_cmpc (str1=0xe8002000b00020, 
    str2=0x10e1518 "audiocpu") at src/lib/util/astring.c:439
439		const char *s1 = str1->text;
(gdb) bt
#0  0x0000000000db8f11 in astring_cmpc (str1=0xe8002000b00020, 
    str2=0x10e1518 "audiocpu") at src/lib/util/astring.c:439
#1  0x00000000007cf818 in memory_region (machine=0x46880c0, 
    name=0x10e1518 "audiocpu") at src/emu/mame.c:840
#2  0x000000000074360d in assemble_common (machine=0x46880c0, 
    cartslot=0x4689420) at src/mess/devices/aescart.c:136
#3  0x00000000007439e6 in assemble_std (machine=0x46880c0, image=0x4689420)
    at src/mess/devices/aescart.c:263
#4  0x0000000000743d35 in device_load_aes_cartridge (image=0x4689420)
    at src/mess/devices/aescart.c:436
#5  0x000000000070c1c4 in device_load_cartslot (image=0x4689420)
    at src/mess/devices/cartslot.c:274
#6  0x0000000000779518 in image_finish_load (device=0x4689420)
    at src/mess/image.c:911
#7  0x000000000077781b in mess_postdevice_init (machine=0x46880c0)
    at src/mess/mess.c:230
#8  0x00000000007d0eb2 in init_machine (machine=0x46880c0)
    at src/emu/mame.c:1546
#9  0x00000000007cebb8 in mame_execute (options=0x4650230)
    at src/emu/mame.c:333
#10 0x0000000000782e8c in cli_execute (argc=6, argv=0x7fff2f41e8c8, 
    osd_options=0x10e2d80) at src/emu/clifront.c:172
#11 0x0000000000744139 in main (argc=6, argv=0x7fff2f41e8c8)
    at src/osd/sdl/sdlmain.c:392


The same segfault occurs when trying Shock Troopers, but admittedly it's not marked as AES (or MVS for that matter, but still).
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/08/09 12:20 AM

Check the region sizes in ng_aes.c - if you overflow one it'll crash very much like that.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 06/08/09 01:03 AM

Ah good, got them all working now. Silly me was only looking at the size of the 68k CPU region :P
Posted By: Firewave

Re: SVN builds - new driver flood - 06/08/09 09:19 PM

We should make "official" XML packs available for the "good" games supported (in favour of the [at least in the past] much discussed static gamelists in the source). Just like Pugsy does for cheats.

Can the XML also be seperated from the rom data? So if you wanna play tpgolf you got the tpgolf.xml in your aes folder and a tpgolf.zip beside where the data can be found.

Would also avoid the distribution of the "official" XMLs with the "illegal" romdata in one archive.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/08/09 09:28 PM

Mizapf bypassed a lot more of the core than I was expecting, so we may or may not get standard behavior with .rpks. Separation from the ROM data doesn't seem possible - at best you could put the xmls in the AES.zip with the *other* illegal rom data.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/09/09 03:57 AM

A little more fun - I added preliminary ATAPI and GD-ROM drive emulation (including sector DMA) and we get the full startup animation with the dot making the letters bounce and then this.

Guess it's time to figure out how controllers work. Or at least how to cheat the NVRAM so it thinks the time's been set ;-)

Posted By: etabeta78

Re: SVN builds - new driver flood - 06/10/09 01:09 PM

Originally Posted By Firewave
We should make "official" XML packs available for the "good" games supported (in favour of the [at least in the past] much discussed static gamelists in the source). Just like Pugsy does for cheats.


yeah, I think xml could be added to Wiki (they only include crc info). Maybe together with a clrmame dat to rebuild the sets from MAME romsets.

static gamelists are parallel to xml and should not be mixed. static gamelists, if ever included, will be a coexisting alternative to the capability of loading any piece of software. the latter, in neogeo, would be impossible without xml layouts

Originally Posted By Firewave
Can the XML also be seperated from the rom data? So if you wanna play tpgolf you got the tpgolf.xml in your aes folder and a tpgolf.zip beside where the data can be found.


I would vote no, but I would agree with the majority. the point of xml support is to replace older headered formats with something cleaner: a zip containing separate chip dumps + xml layout. Neogeo is indeed a borderline case, because MAME already allows to have good dumps, but splitting xml from other data could create problems when we shall work with other systems.

Originally Posted By Firewave
Would also avoid the distribution of the "official" XMLs with the "illegal" romdata in one archive.


who cares if official xmls are spread with illegal data? the important thing is to stress that official website of the project ONLY distributes the legal xml data and no copyrighted items

Originally Posted By R. Belmont
Mizapf bypassed a lot more of the core than I was expecting, so we may or may not get standard behavior with .rpks.


what do you mean for "standard behavior"?


overall, great news about the neogeocd!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/10/09 02:28 PM

By standard behavior I mean things like "roms/folder == roms/folder.zip" and that sort of thing. Also, there is good news about the neogeocd? My post above is Dreamcast smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 06/10/09 02:58 PM

I should have read more carefully blush blush

can I restate: great news about dreamcast!
Posted By: Shideravan

Re: SVN builds - new driver flood - 06/22/09 02:41 AM

I have some questions about the new system craft: for what i know is a homebrew system thats was made with a microcontroller (atmel 8-bit) and adaptations to plug with outputs to show a demo in last year...
1-the driver and dumps was made with approvementnt of creator? did he collaborate with MESS project or craft driver?
2-in the website of lft, we can download the firmware of the system, but why MESS use a bin version of it, unless the original eeprom.raw and flash.hex?...
3-why thats marked as a console?
thanks
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/22/09 03:00 AM

1) lft has no involvement and he freely gives away information on the project. Also, his new demo is on completely different hardware.
2) Emulators need binaries, not hex files.
3) Since it's a bare microcontroller you could claim it's both a computer and a console. Or neither. Since in reality MESS only cares about that flag to know if the device has a keyboard and needs the annoying mode switch enabled calling it a console is correct.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/22/09 08:17 AM

SVN r5052+

FM-7 driver

Xevious (Floppy Disk)
Punch Ball Mario Bros. (Tape)



Now I wait for the FM77AV/40SX support to play games like Space Harrier. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/24/09 07:12 PM

Craft (Linus Akesson) driver

Convert .hex file to .bin file

Example craft.hex and the converter Hex2bin

Hex2bin.exe craft.hex

a file called craft.bin will created.

To check the CRC32 and SHA1 i use HashCalc (Freeware)
Posted By: R. Belmont

Re: SVN builds - new driver flood - 06/26/09 04:58 AM

A little hacking around gets us back (with current code!) to:



And unlike with the old PPC core it's actually trying to boot the ATAPI CD-ROM now. If I could find BeOS install disks we might get someplace.

All is not perfect though - for some reason it only boots if you run -debug. You can immediately dismiss the debugger and it'll boot, but that needs to be fixed. We're working on it smile
(Oh, and the PPC change for this isn't getting commited to SVN quite yet - I want to try and fix the -debug thing first).
Posted By: judge

Re: SVN builds - new driver flood - 06/26/09 05:46 AM

RB: Cool!
Posted By: Justin

Re: SVN builds - new driver flood - 06/26/09 06:16 AM

There are CD images of BeOS R5 out there, I don't know if that will run on the original BeBox hardware or not though.
Posted By: ReadOnly

Re: SVN builds - new driver flood - 06/26/09 11:01 AM

Originally Posted By Heihachi_73
Speaking of Master Systems, can someone test Bank Panic on their end? It seems to be broken on all versions except smsj. CRC of the tested ROM image is B4DFB825; works with Meka's built-in BIOS (unless it's just a 'watch the SEGA logo' before the game starts), and in Dega.

Tested in MESS 0.126 with the v1.3, Hang-on and Alex Kidd BIOS sets, both NTSC and PAL.


Your rom is a bad dump
http://datomatic.no-intro.org/index.php?...11C6&n=0048

good crc32 is 655FB1F4

it is sad to see people keep testing or developping for MESS based on cheesy databases *facepalm*
Posted By: Heihachi_73

Re: SVN builds - new driver flood - 06/26/09 05:37 PM

I kind of figured that out a while ago, MESS is at 0.132 now! wink
Posted By: mizapf

Re: SVN builds - new driver flood - 06/26/09 11:35 PM

I did not follow this thread, so I just noticed that there is something about the rpk handling here.

Originally Posted By R. Belmont
Mizapf bypassed a lot more of the core than I was expecting, so we may or may not get standard behavior with .rpks.
...
By standard behavior I mean things like "roms/folder == roms/folder.zip" and that sort of thing.


Please specify. I still don't get the issue. What do you mean by roms/folder == roms/folder.zip?

I certainly did not intend to bypass anything. So what should have been handled in which different way?

Michael
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/28/09 10:04 AM

SVN r5097
megacd2 driver

megacd_model2_bios_2_00w_e.bin [CRC32: 4d5cb8da] [SHA1: f5f60f03501908962446ee02fc27d98694dd157d]

Are you sure the rom/crc is correct ?

It exist another rom (eu_mcd2_930601):

[CRC32: d344f125] [SHA1: 523b3125fb0ac094e16aa072bc6ccdca22e520e5]

xeye driver

us_xeye_931227.bin [CRC32: 4e18b45c] [SHA1: 615f11b4a81cf957ae55fe99b453c4d8355eab46]

it exist another rom (us_xeye_931227)

[CRC32: 290f8e33] [SHA1: 651f14d5a5e0ecb974a60c0f43b1d2006323fb09]

Quote:
06-28-2009

Thanks to TmEE , the we finally have a WonderMega BIOS dump. This is a dump of the Sega branded WonderMega. Also, TmEE dumped his X'eye, and I found that the release done a while back on X-Cult was a faulty rip. This X'eye BIOS actually functions in an emulator, albeit it is a little slow. Pick them up here.

Actually, TmEE's wondermega dump currently on the wiki page is confirmed bad.

He redumped his X'eye by desoldering and dumping the chip instead of dumping it with his genesis dump program. the crc32 of his redump is 290f8e33 which matches my original dump. So the statement above regarding the dump i did last year being faulty and this xeye bios being more functional is wrong, as TmEEs chip redump matches my dump I released here last year. So should probably edit the x'eye bit out of the news post so not to confuse people.


Posted By: mizapf

Re: SVN builds - new driver flood - 06/28/09 01:51 PM

Referring to the zip file usage with RPK: I think I understand what you meant, but it wasn't me. smile

The zip file access is in multcart.c, and I contributed only some parts of it. Most lines from me are not concerned with reading from a zip file; I just use calls to find a file and to read it.

It's just multcart.c which deals with zip files, so if there are mechanisms in the core to handle folders and zip files in an equivalent way, you should be able to change that without doing any harm to upper layers.

Michael
Posted By: Justin

Re: SVN builds - new driver flood - 06/28/09 05:17 PM

Thanks for the info Anna, I'll correct that.

EDIT: Our 2.00W ROM is marked in the source as a confirmed good dump, I noticed the alternate dump you mention yesterday but I don't know which is right.
Posted By: etabeta78

Re: SVN builds - new driver flood - 06/28/09 05:40 PM

we need elbarto's confirmation. the "confirmed good dump" info came from its website.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/28/09 06:12 PM

SVN r5097

FM-77AV/FM-77AV40SX driver

Good progress for these driver.

Some FM-7 games (Miner2049 first picture) now working. Programs which designed for FM-77AV or FM-77AV40SX machines still not working like Space Harrier (second picture).

Posted By: mahlemiut

Re: SVN builds - new driver flood - 06/29/09 12:55 AM

Yes, FM-77AV support still needs much work. Getting there slowly. Only AV-specific game that runs currently, that I know of, is Tower of Druaga. And it looks rather ugly, too. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/29/09 06:08 AM

Originally Posted By mahlemiut
Yes, FM-77AV support still needs much work. Getting there slowly. Only AV-specific game that runs currently, that I know of, is Tower of Druaga. And it looks rather ugly, too. smile


Anyway, you make a good job. smile
Posted By: ElBarto

Re: SVN builds - new driver flood - 06/29/09 10:41 AM

For the mcd bioses :

the f5f60f03501908962446ee02fc27d98694dd157d one is correct, dumped by me and other people, I just have to update the information on my website.
the 523b3125fb0ac094e16aa072bc6ccdca22e520e5 is a bad dump of the f5f60f03501908962446ee02fc27d98694dd157d, it has shaddowed vectors. It happened when dumping with the megacd active.

The newer x-eye dump is indeed correct, I haven't check myself but you can totally trust TmEE on that.
He dumped the wondermega one too.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 06/29/09 11:18 AM

Originally Posted By ElBarto
For the mcd bioses :

the f5f60f03501908962446ee02fc27d98694dd157d one is correct, dumped by me and other people, I just have to update the information on my website.
the 523b3125fb0ac094e16aa072bc6ccdca22e520e5 is a bad dump of the f5f60f03501908962446ee02fc27d98694dd157d, it has shaddowed vectors. It happened when dumping with the megacd active.

The newer x-eye dump is indeed correct, I haven't check myself but you can totally trust TmEE on that.
He dumped the wondermega one too.


Thanks for the clarification. smile
Posted By: gigadeath

Re: SVN builds - new driver flood - 07/01/09 04:02 PM

As ElBarto said the right CRC for PAL MCD 2.00W is 4D5CB8DA.

Latest dumps from X-Cult:
Wondermega 1.00 JP --> D21FE71D
X'Eye 2.00 US --> 290F8E33

Both dumps comes directly from the chips, not from trasfer cable. Other dumps are bad.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/03/09 10:50 AM

Just will tell your guys, today it is the first time to compile (GCC) MESS for my self. smile

It is working.
Posted By: judge

Re: SVN builds - new driver flood - 07/03/09 10:57 AM

\o/
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/03/09 11:04 AM

As next I will make a private build with Arbee´s WIP GBA driver. smile
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 07/03/09 11:09 AM

When will we see your first bug fix submission wink
Posted By: Heretical_One

Re: SVN builds - new driver flood - 07/03/09 11:11 AM

Probably before I get anywhere with the NES driver frown


(Five minutes later...)
That should teach me to open my big mouth...
Posted By: R. Belmont

Re: SVN builds - new driver flood - 07/03/09 01:45 PM

I've actually done some work on the GBA driver lately (not just making it run under current MESS). Fixed some issues with interrupts but raster interrupts still aren't right. For some reason the games I'm using to test only enable them during vblank, which is obviously wrong smile Also a recently released prototype ROM (Turok) is showing an ARM7 core bug I'm trying to track down.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/03/09 01:53 PM

Originally Posted By R. Belmont
I've actually done some work on the GBA driver lately (not just making it run under current MESS). Fixed some issues with interrupts but raster interrupts still aren't right. For some reason the games I'm using to test only enable them during vblank, which is obviously wrong smile Also a recently released prototype ROM (Turok) is showing an ARM7 core bug I'm trying to track down.


Thank you so much Arbee, for your c l e a r description how to add gba driver source to compile it.
It works. smile
Posted By: mahlemiut

Re: SVN builds - new driver flood - 07/04/09 04:37 AM

Originally Posted By Anna Wu
Some FM-7 games (Miner2049 first picture) now working. Programs which designed for FM-77AV or FM-77AV40SX machines still not working like Space Harrier (second picture).


You might want to take another look at Space Harrier now... wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/04/09 10:33 AM

Originally Posted By mahlemiut
Originally Posted By Anna Wu
Some FM-7 games (Miner2049 first picture) now working. Programs which designed for FM-77AV or FM-77AV40SX machines still not working like Space Harrier (second picture).


You might want to take another look at Space Harrier now... wink


Thanks Barry, can now start and play Space Harrier. The graphic (colors/sprites etc.) need some fixes. smile

Screenshot (Animated GIF)



Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/04/09 11:44 AM

SVN r5140

TIKI 100 driver

Floppy Disk
TIKI Invaders

Posted By: Tomas K

Re: SVN builds - new driver flood - 07/04/09 04:56 PM

Originally Posted By R. Belmont
A little hacking around gets us back (with current code!) to:

And unlike with the old PPC core it's actually trying to boot the ATAPI CD-ROM now. If I could find BeOS install disks we might get someplace.


Excellent work! cool I've just made disk images of BeOS DR 8.1 that came bundled with my BeBox, also R5 which were the last release with PPC support.


Posted By: R. Belmont

Re: SVN builds - new driver flood - 07/04/09 05:00 PM

Awesome. PM me about the images and I'll hook up the CD-ROM drive and see what happens smile
Posted By: mahlemiut

Re: SVN builds - new driver flood - 07/05/09 01:15 AM

Originally Posted By Anna Wu
Thanks Barry, can now start and play Space Harrier. The graphic (colors/sprites etc.) need some fixes. smile


Yeah, still need to do the line drawing part of the ALU, and also handling the masking bits. Will get there eventually. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 07/05/09 01:38 AM

Considering I'd never heard of the FM7 about 6 weeks ago I think the progress is just fine smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/07/09 12:41 PM

Einstein TC-01 driver

With the new SVN builds, I am not able to load Floppy Disks.

As example xbasic.dsk (Xtal Basic)

With the last official version i can load.

Screenshot 1 (last official release)
Screenshot 2 (SVN r5171, tried also bobz SVN r5165 build)

Posted By: Duke

Re: SVN builds - new driver flood - 07/07/09 01:22 PM

Yes, this was caused by napobears changes. He said he will look into it, but it looks like he hasn't had time yet.
Posted By: napobear

Re: SVN builds - new driver flood - 07/07/09 03:11 PM

I'm trying to understand the cause. I have to understand the real machine first. Someone can help me?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/07/09 07:15 PM

Understand, so I will use the older driver.
Posted By: napobear

Re: SVN builds - new driver flood - 07/07/09 11:20 PM

Fixed.
Already sent the patch to npwoods.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/08/09 08:30 AM

Originally Posted By mahlemiut
Originally Posted By Anna Wu
Thanks Barry, can now start and play Space Harrier. The graphic (colors/sprites etc.) need some fixes. smile


Yeah, still need to do the line drawing part of the ALU, and also handling the masking bits. Will get there eventually. smile


SVN r5179

Looks better and better.

Screenshot (Animated GIF)



Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/08/09 08:32 AM

Originally Posted By napobear
Fixed.
Already sent the patch to npwoods.


Nice !
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/08/09 04:59 PM

SVN r5182

Einstein TC-01 driver

Game: Jet Set Willy (Floppy Disk)

Screenshot (Animated GIF)




Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/09/09 03:49 PM

Jupiter ACE driver

Is it possible to load and start any software (Tape/Cartridge) successful ?

If I try to load a tape, I get a crash.
Cartridges i can load but then the emulation freeze. The emulated keyboard is blocked.
Posted By: Robbbert

Re: SVN builds - new driver flood - 07/09/09 07:51 PM

I've never had any luck with that either.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/10/09 07:52 AM

Maybe the status shall be changed.

Posted By: etabeta78

Re: SVN builds - new driver flood - 07/10/09 08:11 AM

I seem to remember those cassettes were working when Judge converted cassettes to devices... but he can probably be more precise
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/10/09 02:46 PM

r5203

Just update SVN. After compilation, i try to start mess.exe and get the error :
Invalid token 62 in machine config

Something is broken ?
Posted By: Duke

Re: SVN builds - new driver flood - 07/10/09 02:49 PM

Did you make a clean build?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/10/09 02:53 PM

Ups, this can be the reason. After an MAME core update, I need a full clean build ?
That means, I have to remove all in the obj folder ?
Posted By: Duke

Re: SVN builds - new driver flood - 07/10/09 03:04 PM

Yes, either remove the obj folder or enter your usual compile line and add "clean" at the end (for example, "make TARGET=mess clean"). After a core update, a clean build is almost always needed, but it can be needed with smaller changes too.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/10/09 03:07 PM

Originally Posted By Duke
Yes, either remove the obj folder or enter your usual compile line and add "clean" at the end (for example, "make TARGET=mess clean"). After a core update, a clean build is almost always needed, but it can be needed with smaller changes too.


Thank you so much for your clear words. I will test after the dinner. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/10/09 11:40 PM

Now it works. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/14/09 07:31 AM

SVN r5211
trs80m3 driver

The driver with selected Level 2 Rev. B/C bios is not working right ?

Someone is able to load and start any disk with Level 1 or Level 2 Network III bios ?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/14/09 09:48 PM

SVN r5205+

Virtual Boy driver (still in progress)

Posted By: etabeta78

Re: SVN builds - new driver flood - 07/14/09 10:29 PM

even better screens here

http://micko-wip.blogspot.com/

congratulations to micko!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/14/09 10:46 PM

SVN r5230+

Visual 1050 driver

If i mount/load a unzipped .dsk file, no problems.
With zipped .dsk file I get a error.

Screenshot (Animated GIF)



Posted By: etabeta78

Re: SVN builds - new driver flood - 07/14/09 10:55 PM

are there multiple files in the zip file? also try to unzip and then rezip the image, sometimes it's a non-standard zipfile image to create problems
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/14/09 11:00 PM

Originally Posted By etabeta78
are there multiple files in the zip file? also try to unzip and then rezip the image, sometimes it's a non-standard zipfile image to create problems


No, the zipped file content only ONE .dsk file.
Example: cpm30a.dsk to cpm30a.zip
Also I rezipped the file several times.
If the zipped file is wrong, the disk will not boot.
It is booting but later come the error.
Posted By: etabeta78

Re: SVN builds - new driver flood - 07/14/09 11:02 PM

it's been a long time since I've seen a similar problem... zip support in MESS is usually very stable. not sure about the way disks are handled by the driver: probably Curt may give more info/advices
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/14/09 11:05 PM

Originally Posted By etabeta78
it's been a long time since I've seen a similar problem... zip support in MESS is usually very stable. not sure about the way disks are handled by the driver: probably Curt may give more info/advices


This is the reason why I post it. smile
I am sure, Curt read it.
Posted By: Justin

Re: SVN builds - new driver flood - 07/15/09 01:25 AM

Notice the error is a write protect error - MESS is not capable of writing to disks inside of zip archives (because it would have to re-zip them every time) so zipped disks behave like write-protected disks. If the emulated computer doesn't like write-protected disks for whatever reason, then you'll just have to use unzipped images. (You might want to keep unmodified backups somewhere though.)
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/15/09 06:59 AM

Originally Posted By Justin
Notice the error is a write protect error - MESS is not capable of writing to disks inside of zip archives (because it would have to re-zip them every time) so zipped disks behave like write-protected disks. If the emulated computer doesn't like write-protected disks for whatever reason, then you'll just have to use unzipped images. (You might want to keep unmodified backups somewhere though.)


This can be the reason. Thank you for the Info !
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/15/09 10:47 PM

Last SVN builds.

abc80/abc806/abc800m driver

If i start the emulation, I get a crash.
I informed Curt already.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/16/09 07:01 AM

Originally Posted By Anna Wu
Last SVN builds.

abc80/abc806/abc800m driver

If i start the emulation, I get a crash.
I informed Curt already.


Problem solved after new clean build.
Conclusion, before I post a problem I will make a clean build to test again.

Thanks Curt !
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/20/09 08:03 PM

It seems the tape function for Sord M5 is broken since the version 0.116.

Just test the tool " castool " to convert the .cas (for Sord M5) to .wav format. It is also not working.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/21/09 04:18 PM

SVN r5286

PK8020 Korvet
Lode Runner (Floppy Disk)



Thanks Micko !
Posted By: Anna Wu

Re: SVN builds - new driver flood - 07/25/09 10:27 AM

Originally Posted By Anna Wu
It seems the tape function for Sord M5 is broken since the version 0.116.

Just test the tool " castool " to convert the .cas (for Sord M5) to .wav format. It is also not working.


It seems castool is working (cannot test in practice), my parameter was wrong.
But the loading tape problem still exist.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 07/26/09 06:29 AM

How's the ARM9 work that went into MAME recently?

About this good ;-)

Posted By: Heretical_One

Re: SVN builds - new driver flood - 07/26/09 06:48 AM

... will it play any of the Korean Snes9x hacks? smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 07/26/09 07:25 AM

I'm aiming for being able to run PSX4GP2X on it and thereby achieving commercial PSX games on MESS before the actual PSX driver does ;-)
Posted By: etabeta78

Re: SVN builds - new driver flood - 07/26/09 11:54 AM

coooool!!!!!!!!!!!!!!!!!!!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 07/26/09 03:44 PM

A little tweek (never do RGB format conversion at 2 AM):

Posted By: Justin

Re: SVN builds - new driver flood - 07/26/09 05:38 PM

I hope there wasn't anything important in that fine print smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/01/09 10:55 AM

SVN r5332

APF Imagination Machine

I am not able to load successful any tapes.
The tape is playing but ignore the end of tape time.
Also have problems with the floppy disks, can not read content.

Posted By: ranger_lennier

Re: SVN builds - new driver flood - 08/01/09 06:27 PM

Did this ever work? APF emulation is pretty spotty, as I recall.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/01/09 08:08 PM

Really, not know.

sysinfo: Cassette loading and saving is working
Posted By: Robbbert

Re: SVN builds - new driver flood - 08/01/09 10:11 PM

Tape loading never worked much for me either. It would usually lose sync partway through, and load gibberish. Never had a disk to test with. The computer should probably be marked as not working.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/02/09 09:00 AM

SVN r5336
Socrates Educational Video System

Screenshot (animated GIF)

Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 08/02/09 06:47 PM

Anna: super painter is now usable (keyboard arrow keys move cursor and left alt for the trigger/button; also try keypad 8/4/6/2 and keypad enter for faster movement; the movement is way too fast in both modes due to lack of waitstates though)

LN
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/02/09 06:53 PM

Thanks, I will test again. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/02/09 07:14 PM

SVN r5338
Socrates Educational Video System

Just select/copy/paste function in Super Painter tested smile



Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/04/09 09:05 PM

SVN r5342+

Atari Jaguar
Atomix (load to $4000)

Thanks for the Quickload support, robbbert. smile

Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/05/09 11:44 AM

Just a reminder.
The tape support for the Nascom drivers is still broken.
Also to have problems (Nascom 2) to load floppy disks.
After start, it freeze or make reset.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/07/09 02:51 PM

SVN r5367+
PCW8256
System boot now CP/M

Thanks Barry. smile


Posted By: mahlemiut

Re: SVN builds - new driver flood - 08/07/09 10:00 PM

Ooo, I didn't know there was a PCW version of Tau Ceti. laugh
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/08/09 02:14 PM

SVN r5355+
Junior Computer (Elektor)

Thanks Stefano and Manfred. smile



Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/08/09 04:41 PM

SVN r5372+
Pyldin 601

At moment, no device like Floppy Disk support exist.
Is it possible to support Paste/Copy (Edit > Paste) function to use exist basic listings ?
The Paste function is not active.
Posted By: etabeta78

Re: SVN builds - new driver flood - 08/08/09 08:23 PM

yes, I think that "Paste" needs natural keyboard emulation to work and Pyldin 601 still does not have it. I will fixed it soon
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/08/09 08:32 PM

Thanks ! smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/09/09 07:30 AM

SVN r5377+
GBA

The driver show me twice in MESS GUI.

Edit: Found the reason in messdriv.c smile

DRIVER( gba ) /* Nintendo Game Boy Advance */
DRIVER( snes ) /* Nintendo Super Nintendo NTSC */
DRIVER( snespal ) /* Nintendo Super Nintendo PAL */
DRIVER( sfcbox ) /* Nintendo Super Famicom Box */
DRIVER( n64 ) /* Nintendo N64 */
DRIVER( pokemini ) /* Nintendo Pokemon Mini */
DRIVER( gba ) /* Nintendo Game Boy Advance Handheld */

It seems, I forget to remove the entry which I add by hand in the past.
Posted By: etabeta78

Re: SVN builds - new driver flood - 08/09/09 07:36 AM

it's only in your build. you probably had the driver enabled when it was WIP, and you forgot to remove the line now that it's officially supported wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/09/09 07:38 AM

Yes, you are right. See my previous posting. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 08/09/09 07:41 AM

ah ok, you had already found out. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 08/09/09 08:55 AM

Originally Posted By Anna Wu
Is it possible to support Paste/Copy (Edit > Paste) function to use exist basic listings ?
The Paste function is not active.


Paste should work now in pyl601 (svn 5382)
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/09/09 09:52 AM

Thanks again. Unfortunately, a correct insertion does not work. Maybe it is a problem of the driver emulation.
So better I wait ...

Example: 10 SCREEN3:CALL$13:IFPEEK($ed41)<>3THEN10

On the screenshot, you will see the missing strings.

Posted By: Tafoid

Re: SVN builds - new driver flood - 08/09/09 12:39 PM

Originally Posted By etabeta78
it's only in your build. you probably had the driver enabled when it was WIP, and you forgot to remove the line now that it's officially supported wink


Nice to see it officially added. Though, I would suggest that until more work can be done on it - it would be safer to include a GAME_IMPERFECT_GRAPHICS flag. The 4 games I tried all had issues related to video.. some of which make them quite unplayable.
Posted By: etabeta78

Re: SVN builds - new driver flood - 08/09/09 02:38 PM

I agree. Can I go on and add the flag, Arbee?

also, is the rom name really supposed to be "gba_state->bin"?!? quite weird, in my opinions...
Posted By: R. Belmont

Re: SVN builds - new driver flood - 08/09/09 03:05 PM

I'll add the flag. If you'd like to add windows or scaling/rotating sprites or color add/subtract feel free though (the video h/w is a pretty obvious extension of the SNES).

BTW, Jaguar should have separate PAL/NTSC drivers instead of a dip switch.
Posted By: etabeta78

Re: SVN builds - new driver flood - 08/09/09 03:16 PM

let me stick to SNES for the time being... I'll take a look at GBA later (after discussing with you what to look at wink )
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/09/09 06:34 PM

Originally Posted By Anna Wu
SVN r5372+
Pyldin 601

At moment, no device like Floppy Disk support exist.
Is it possible to support Paste/Copy (Edit > Paste) function to use exist basic listings ?
The Paste function is not active.


SVN r5384+
Pyldin 601
Floppy Disk device supported

Very nice, Micko smile


Posted By: R. Belmont

Re: SVN builds - new driver flood - 08/10/09 12:49 AM

Rev 5391 adds the anti-alias filtered speaker toggle emulation that was submitted to MAMEdev in May and ignored because it was MESS-only-in-common-code and nobody spoke up for it smile

It's a noticable difference and I also hooked it up to the Apple II series where it sounds much better and fixes a long-time issue with the mixing on the IIgs (the DC offset from the raw unfiltered speaker toggle was reducing the available headroom for the ES5503).
Posted By: Justin

Re: SVN builds - new driver flood - 08/10/09 05:20 AM

FWIW, MSVC spits this out:
c:\games\emulated\mess\src\emu\sound\speaker.c(243) : warning C4701: potentially uninitialized local variable 'sampled_time' used
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/10/09 06:28 AM

SVN r5393+
Socrates Educational Video System
Support now cartridges as selectable roms.

Thanks Lord Nightmare !




Posted By: R. Belmont

Re: SVN builds - new driver flood - 08/10/09 03:37 PM

Justin: what version? Aaron says it's fine on his MSVC (I think he uses either '05 or '08).
Posted By: Justin

Re: SVN builds - new driver flood - 08/11/09 04:54 AM

2008 Express Edition
Posted By: Curt Coder

Re: SVN builds - new driver flood - 08/13/09 09:21 AM

Originally Posted By Anna Wu
SVN r5367+
PCW8256
System boot now CP/M


This also breaks Bondwell 12/14.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/13/09 10:22 AM

Originally Posted By Curt Coder
Originally Posted By Anna Wu
SVN r5367+
PCW8256
System boot now CP/M


This also breaks Bondwell 12/14.


Maybe the nec765.c can be fixed, so both systems have no booting conflicts ?
Posted By: mahlemiut

Re: SVN builds - new driver flood - 08/13/09 10:43 AM

Originally Posted By Curt Coder
This also breaks Bondwell 12/14.

All the change to nec765.c should do is make the Recalibrate command actually take the time it is supposed to before completion. There was one other change I've made since, that makes Recalibrate take time to perform when it is already at track 0, again, simply because the PCW does not like it finishing immediately.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 08/14/09 03:28 AM

Ok, from taking a quick look at the Bondwell 12/14 boot process, it looks to me as though it actually ought to wait for the Recalibrate/Seek function to end before it goes and does Sense Interrupt Status.

So, I've added code to set the execution mode bit of the main status register during a seek or recalibrate function, and moved the nec765_idle() call from nec765_seek_setup() to nec765_seek_complete(), as nec765_idle() clears the execution mode bit.

If this makes sense to everyone, I'll go and commit it to SVN. I have no software for the Bondwell, so I can't test it any further than it complaining after wanting a system disk. The PCW driver still works with this change, and I also quickly tested the CPC and X68000 drivers, just in case, which also seem to still work.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 08/14/09 06:50 AM

I've commited the small change I made to the upd765. I can confirm that Bondwell 12/14 will boot CP/M now, without breaking the PCW. smile Seems like a rather slow system however, although it would to be a similar speed in 0.133 also. Funky speech program on the disk, though. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/14/09 08:08 AM

SVN r5430
mtx500/mtx512/rs128 driver

I am not able to load a snapshot. Not sure, i make some mistake.

Posted By: Duke

Re: SVN builds - new driver flood - 08/14/09 11:09 AM

Snapshot loading is not working in those drivers.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/14/09 11:24 AM

Originally Posted By Duke
Snapshot loading is not working in those drivers.


So I can not load any software ?
Posted By: Duke

Re: SVN builds - new driver flood - 08/14/09 03:18 PM

Nope, you can't yet. Adding cassette support should be fairly easy, but there are no real docs and those things always go over my head smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 08/15/09 02:55 AM

FYI: SVN #5436 has a major revision to the YM2612/3834 emulation containing the latest SpritesMind research (or at least the latest that Eke-Eke incorporated into his version). If you had any Genesis/Megadrive games that sounded problematic previously, do try them again. The standard GEMS tortures like Scooby-Doo Mystery sound terrific now.
Posted By: Kale

Re: SVN builds - new driver flood - 08/15/09 10:11 AM

Have you ever tried to check the cause about why the Micro Machines 2/'96 clacson sound (by pressing button C) never sounded right?
Posted By: Haze

Re: SVN builds - new driver flood - 08/15/09 02:55 PM

compared to a few days ago the stage introduction music in aladdin (aladbl in MAME) sounds like it's regressed.

One of the bells appears to get cut short / distorted now compared to the previous build. Should probably be verified.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 08/15/09 02:57 PM

They cited Aladdin specifically as something this addressed. I'll see about getting a real sample.
Posted By: Justin

Re: SVN builds - new driver flood - 08/15/09 06:31 PM

Zero Wing sounds awesome now smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/20/09 09:35 PM

SVN r5470
Virtual Boy

Reality Boy Demo 1 (PD)
Button-Test is working



Posted By: Shideravan

Re: SVN builds - new driver flood - 08/20/09 09:55 PM

Great!
By I think thats is a little strange thats the default controller " right digital pad" is attributed to same buttons as the A,B,L and R.
But, no problems...
It's just to configure!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/26/09 09:02 AM

SVN r5500

Atari Jaguar driver

Just check the Quickload support again.
I tested some Demos and see the graphics are bigger so not all more visible. Before it was not perfect but ok.



Posted By: Robbbert

Re: SVN builds - new driver flood - 08/26/09 10:54 AM

Works for me... (for some reason the screenshot is stretched, although the game looks normal on my monitor)..

Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/26/09 11:24 AM

Strange

Here a original screenshot maked by MESS



Will check my settings again.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/26/09 12:01 PM

I get not your screen if I load demos by using Quickload device. I tried different settings (aspect ratio, view, screen etc.)
No different versions of these demos exist. So I think we use the same demos.

It seems it is not a general problem, because if load some carts (not Quickload images with start address $4000), the screen looks ok (see screenshot)

Not understand frown

Posted By: Shideravan

Re: SVN builds - new driver flood - 08/26/09 03:16 PM

It's normal!
It's from UI...
When you take a scren at the deafult configuration, you'll get the equivalent for the windows resolution.
You can change it in Options>Default system options>Miscelaneous> snapshot view!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/26/09 04:24 PM

Originally Posted By Shideravan
It's normal!
It's from UI...
When you take a scren at the deafult configuration, you'll get the equivalent for the windows resolution.
You can change it in Options>Default system options>Miscelaneous> snapshot view!


Sorry, this is not normal. I was talking about screen view on monitor, not about the snapshot view.
Test the game " Atomic " (not Atomic Reloaded) and tell me your result. This image is distributed under the terms of the GPL License and use the start address $4000 which you need to load for the Quickload device.
Posted By: Shideravan

Re: SVN builds - new driver flood - 08/26/09 04:46 PM

SVN r5503

Ok, I tested Atomic and a had a half screen too...
I seens that is not Anna configuration...

But Zool is normal.
Well, a little regression in graphics on the upper screen that show some garbage when you fall from the plataforms, but the screen is normal...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/26/09 04:50 PM

Originally Posted By Shideravan
SVN r5503

Ok, I tested Atomic and a had a half screen too...
I seens that is not Anna configuration...

But Zool is normal.
Well, a little regression in graphics on the upper screen that show some garbage when you fall from the plataforms, but the screen is normal...


Yes, I wrote this already.
The problem occurs, if you load images on Quickload device.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/26/09 04:57 PM

You can test another demo called Osmozys.

Posted By: Haze

Re: SVN builds - new driver flood - 08/26/09 05:55 PM

I don't know much about the Jaguar, but is it missing some blend modes, or are the bubbles in Fishing Frenzy (cojag) really meant to be that ugly?
Posted By: Robbbert

Re: SVN builds - new driver flood - 08/26/09 10:45 PM

They are rather ugly aren't they.. checked against 0.130 and is the same. I could investigate if you like.

Osmozys runs ok, although I think there might be some things not showing. Tried in Project Tempest, and the twirling balls don't show at all.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/26/09 11:56 PM

Originally Posted By robbbert
They are rather ugly aren't they.. checked against 0.130 and is the same. I could investigate if you like.

Osmozys runs ok, although I think there might be some things not showing. Tried in Project Tempest, and the twirling balls don't show at all.


SVN r5503
Osmozys demo

It looks not normal on my monitor, still not all visible
Maybe we have different versions of the jaguar driver.

First screenshot (MESS SVN r5503), second screenshot (MESS SVN r5397), third screenshot (PT)



Posted By: Robbbert

Re: SVN builds - new driver flood - 08/27/09 09:23 AM

You have to wait for the two title screens to be done, after that you can see the balls.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/27/09 09:34 AM

Originally Posted By robbbert
You have to wait for the two title screens to be done, after that you can see the balls.


It seems we have a understanding problem. I stop the discussion now.
Posted By: Kale

Re: SVN builds - new driver flood - 08/27/09 12:47 PM

r5507 /src/mess/drivers/x1.c: [X1]: Added preliminary bitmap support

Starts to show up things nicely...

http://mamedev.emulab.it/kale/fast/files/0060_1148835870.png
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/27/09 01:51 PM

Originally Posted By Kale
r5507 /src/mess/drivers/x1.c: [X1]: Added preliminary bitmap support

Starts to show up things nicely...

http://mamedev.emulab.it/kale/fast/files/0060_1148835870.png


Nice smile

Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/28/09 09:02 PM

Lord Nightmare, can you take a look, please ?

Just tried to compile the last SVN r5516.
Error:

Compiling src/mess/drivers/dectalk.c...
cc1.exe: warnings being treated as errors
src/mess/drivers/dectalk.c:235: warning: initialization from incompatible pointer type
mingw32-make: *** [obj/windows/mess/mess/drivers/dectalk.o] Error 1

Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/28/09 10:49 PM

Thanks for the fix, Lord Nightmare. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/28/09 11:12 PM

SVN r5514
Sharp X1 driver
Hopefully fixed the colors minus one small bug
Game: Super Mario Bros. Special

Kale, the sky is black instead of blue.






Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/29/09 09:36 PM

SVN r5529
Sharp X1 driver

Some new screenshots (animated GIF)

Posted By: Kale

Re: SVN builds - new driver flood - 08/29/09 09:40 PM

Originally Posted By Anna Wu
SVN r5514
Sharp X1 driver
Hopefully fixed the colors minus one small bug
Game: Super Mario Bros. Special

Kale, the sky is black instead of blue.



It accesses the DMA register, the X-1 DMA is probably the most awkward DMA that I've ever seen (works with a single byte register!). I'll support it at some point but don't hold your breath wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 08/31/09 08:52 PM

SVN r5548
Mikron-2 driver
Game: Pacman

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/01/09 11:03 PM

SVN r5557
Sharp X1 Turbo driver
Game : Diablo, Spy vs Spy

Posted By: Shideravan

Re: SVN builds - new driver flood - 09/04/09 12:29 AM

SVN r5571

Virtual boy
Game: Virtual Pong (PD)



Thanks!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/04/09 05:53 AM

SVN r5573
Sharp X1 driver

Thanks for improved keyboard support.
I am now able to use my game collection disk´s with menu selection. smile

Posted By: Kale

Re: SVN builds - new driver flood - 09/05/09 02:21 PM

r5592 /src/mess/drivers/x1.c: [X1]: fixed PCG-based colors

Actually a rather stupid bug, I was using the wrong color bank for whatever reason...

http://mamedev.emulab.it/kale/fast/files/0304.png

http://mamedev.emulab.it/kale/fast/files/0306.png

http://mamedev.emulab.it/kale/fast/files/0307.png

http://mamedev.emulab.it/kale/fast/files/0308_909981393.png

(and yes, the black pen thing in Super Mario Bros. SP is now fixed wink )
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 04:15 PM

Good news, thanks. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/05/09 04:30 PM

Nice. Looks like I'm gonna have to Google up some X1 games smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 06:16 PM

I wish I can play HYDLIDE 3 with X1 Turbo driver, a RPG game.



Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 08:16 PM

Just a question, do you like more if i make screenshots with internal snapshot function of MESS ?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/05/09 08:17 PM

They're a bit smaller, but I don't think anyone cares either way smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 08:22 PM

The reason why i use my way is :

1.) WYSIWYG
2.) People can see which binary and driver I use.
Posted By: Justin

Re: SVN builds - new driver flood - 09/05/09 08:30 PM

Personally I would prefer internal screenshots because then I can upload them to the wiki smile

FWIW, internal screenshots all record which version of MESS and driver they were taken with, you can view this at the end of the file with a hex editor or with the image viewer on the wiki.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 08:33 PM

OK, no problem. smile
Remember me, if i forget it.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 08:41 PM

Snapshot (Sharp X1 driver) format ok ?



Posted By: Justin

Re: SVN builds - new driver flood - 09/05/09 08:44 PM

Perfect smile

http://mess.redump.net/_detail/mess:drivers:x1:x1:0000m.png
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 09:01 PM



Thanks. smile

Do you wish a specific name for the snapshot file ?
Posted By: Justin

Re: SVN builds - new driver flood - 09/05/09 09:03 PM

Could be handy to have the game title in there if it's not obvious, but other than that it doesn't really matter much.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 09:07 PM

Sharp X1 driver
Game : Space Harrier

Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/05/09 09:09 PM

How do you control Space Harrier? I was able to shoot but not move the man smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/05/09 09:18 PM

Originally Posted By R. Belmont
How do you control Space Harrier? I was able to shoot but not move the man smile


Config Joystick (Options > Joysticks > Joystick 1)
Start the game and go in the menu (Title Screen) and select Joystick instead of Keyboard.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 09/06/09 01:32 AM

Originally Posted By Anna Wu
Originally Posted By R. Belmont
How do you control Space Harrier? I was able to shoot but not move the man smile


Config Joystick (Options > Joysticks > Joystick 1)
Start the game and go in the menu (Title Screen) and select Joystick instead of Keyboard.


You may now use the keyboard controls on Space Harrier now. I kinda used the wrong input port bits for the keypad 'game keys' (were off by one bit).
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/06/09 04:55 AM

Awesome, thanks Barry! smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/06/09 05:27 AM

Originally Posted By mahlemiut
Originally Posted By Anna Wu
Originally Posted By R. Belmont
How do you control Space Harrier? I was able to shoot but not move the man smile


Config Joystick (Options > Joysticks > Joystick 1)
Start the game and go in the menu (Title Screen) and select Joystick instead of Keyboard.


You may now use the keyboard controls on Space Harrier now. I kinda used the wrong input port bits for the keypad 'game keys' (were off by one bit).


Thanks, now I have different control possibilities. smile
Posted By: Kale

Re: SVN builds - new driver flood - 09/06/09 11:11 AM

Originally Posted By Anna Wu
I wish I can play HYDLIDE 3 with X1 Turbo driver, a RPG game.



It's a T&E Soft game, and T&E Soft games generally abuses the console/computer HWs. Unsuprisingly, all three Hydlide behaves in unexpected and broken fashion, *shrug*...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/06/09 11:17 AM

OK, I can not have all. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/08/09 07:50 AM

SVN r5636
PasoGo (KOEI)driver with artwork
Game : GO



PS: ImageShack renamed the file name.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/08/09 08:33 AM

For the gallery, Justin.

4004 Nixie Clock

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/08/09 12:49 PM

Cosmac VIP (VP-711)

Not really working for me.
If this is true, maybe we shall change the status.

Emulation: Good
Color: Good
Sound: Imperfect
Graphics: Good
Save State: Supported

PS: Cosmac VIP (VP-111) is also not working.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/08/09 12:57 PM

Hah, that Nixie Clock is awesome.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 09/08/09 01:42 PM

Originally Posted By Anna Wu
Cosmac VIP (VP-711)
Not really working for me.
PS: Cosmac VIP (VP-111) is also not working.


Hold C key and press R key twice to boot.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/08/09 02:07 PM

Thanks Curt, this was the secret. smile

PS: You have a PM.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/08/09 05:46 PM

SVN r5647
Cosmac VIP (VP-711) driver
Game : Breakout (Quickload)

Just a screenshot.


Posted By: JoJo

Re: SVN builds - new driver flood - 09/08/09 09:13 PM

Originally Posted By Anna Wu
For the gallery, Justin.

4004 Nixie Clock



How do you set/start the clock?!? Keys "1" and "2" are unresponsive on my SDLMESS build...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/08/09 09:29 PM

Originally Posted By JoJo
Originally Posted By Anna Wu
For the gallery, Justin.

4004 Nixie Clock



How do you set/start the clock?!? Keys "1" and "2" are unresponsive on my SDLMESS build...


With the keys 1/2 on my MESS Windows binary build.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 09/08/09 11:06 PM

Works for me, not that I entirely understand exactly how it all works, the buttons do do something, though. (SDLMESS current SVN, Ubuntu 8.04 x86_64)
Posted By: JoJo

Re: SVN builds - new driver flood - 09/09/09 10:58 PM

Definitely there must be something wrong: when I run the driver all I get are the four dots flashing and no digits in the nixie tubes. Keys "1" and "2" do nothing. (SDLMESS current, Gentoo x86_64)

BTW could you check this: start any driver, enter "Input (this machine)" and redefine any key, then press DEL to reset the definition to default - Segfault!
Posted By: mahlemiut

Re: SVN builds - new driver flood - 09/10/09 12:07 AM

Originally Posted By JoJo
BTW could you check this: start any driver, enter "Input (this machine)" and redefine any key, then press DEL to reset the definition to default - Segfault!


No segfault here, it just sets Delete to that input. What it does not do, though, is stop adding extra inputs. ie: When you select an input, enter a key, move to another input, then back again, it adds another input instead of clearing it and showing only the key just pressed. It should only add keys if you *don't* move to another input.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/10/09 05:55 AM

Originally Posted By JoJo
Definitely there must be something wrong:


Sorry guys, on SDLMESS build (Gentoo x86_64) I can not test the 4004 Nixie Clock driver.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/10/09 06:12 AM

SVN r5664
DECTalk DTC-01 (dectalk driver)

@Lord Nightmare

" Using factory settings " smile

PS: Can not record speech synthesis.
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 09/10/09 06:30 AM

Anna Wu: Yes, it doesn't have any input method yet, I'm working on that still. Eventually you should be able to either dump textfiles in it somehow or at least blindly type at it; I need to whip up a sort of 'virtual rs232 terminal' for it.
Or it could be attached to the vt100, too, if we ever get that working.

LN
Posted By: JoJo

Re: SVN builds - new driver flood - 09/10/09 08:19 AM

Originally Posted By mahlemiut
Originally Posted By JoJo
BTW could you check this: start any driver, enter "Input (this machine)" and redefine any key, then press DEL to reset the definition to default - Segfault!


No segfault here, it just sets Delete to that input. What it does not do, though, is stop adding extra inputs. ie: When you select an input, enter a key, move to another input, then back again, it adds another input instead of clearing it and showing only the key just pressed. It should only add keys if you *don't* move to another input.


Which version of SDL are you using? Mine is 1.2.13. There could be another possibility: usually ArBee commits sdl changes only when SVN compilation breaks - what if he forgot to commit a minor change?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/10/09 08:20 AM

Originally Posted By Lord Nightmare
Anna Wu: Yes, it doesn't have any input method yet, I'm working on that still. Eventually you should be able to either dump textfiles in it somehow or at least blindly type at it; I need to whip up a sort of 'virtual rs232 terminal' for it.
Or it could be attached to the vt100, too, if we ever get that working.

LN


Thank you for your explanation. smile

I know, the input is not supported yet.

What i mean with

" Using factory settings "

PS: Can not record speech synthesis.

is, I can not demonstrate the speech synthesis like " Using factory settings ", because I not have a recorded sound file.

You know what I mean ?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/10/09 08:36 AM

SVN r5666
snes driver
Game : SD Gundam GX (J)

Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 09/10/09 05:05 PM

Anna: no, it says on startup:

Dectalk version 2.0 is running
NVR Fault, using factory settings


The reason it says this is because the NVRam default image is bad. I'm still working on fixing that, because unlike most systems, if the nvram is bad the dectalk does NOT rewrite it by default; you have to actually send it a few commands to force it to rewrite the nvram contents.

LN
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/11/09 10:11 AM

SVN r5685
Bondwell 2 (bw2 driver)
WordStar (Floppy Disk)

Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 09/11/09 04:56 PM

BTW: fixed the nvram error in dectalk (sort of; in reality you should have to type a command in to reinitialize it if its clear; I also haven't figured out how to attach it to the generic nvram system, or if its even possible to do so, since the nvram addresses are interleaved with the led access addresses using UDS/LDS)

I'm working on an input system which should allow blindly typing random stuff and having it spoken.

Anna: you mentioned some problem with recording wavefiles from dectalk? can you explain?

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/11/09 05:00 PM

Handling the interleave is trivial.

AM_RANGE(start, end) AM_READWRITE8(nvram_8bit_r, nvram_8bit_w, 0x00ff)
AM_RANGE(start, end) AM_READWRITE8(led_8bit_r, led_8bit_w, 0xff00)

Then both handlers are called as if it's a linear array of bytes.
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 09/11/09 05:04 PM

ok, how the hell did I not figure that one out...
gimme a few minutes to implement that.

Edit: arg, is there an 8-bit-with-mask version of AM_RAM?
so i could do something like:
Code:
AM_RANGE(0x094000, 0x0941ff) AM_WRITE8(led_write, 0x00FF) AM_MIRROR(0x763C00) /* LED array */
AM_RANGE(0x094000, 0x0941ff) AM_RAM8(0xFF00) AM_BASE(&generic_nvram) AM_SIZE(&generic_nvram_size) AM_MIRROR(0x763C00) /* Xicor X2212 NVRAM */


Even that isn't optimal, the best would be if there was some way to access the nvram from within a write function, since the nvram acts as normal sram from the memory map, and is only stored to the non volatile 'backing store' when a different address is written to (0x94200) and is read from the 'backing store' when that same address is read from.

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/11/09 05:05 PM

To be fair it's a relatively new core feature, but extremely handy in situations like this.
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 09/11/09 05:29 PM

see prev message edit
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/11/09 06:29 PM

Originally Posted By Lord Nightmare
BTW: fixed the nvram error in dectalk (sort of; in reality you should have to type a command in to reinitialize it if its clear; I also haven't figured out how to attach it to the generic nvram system, or if its even possible to do so, since the nvram addresses are interleaved with the led access addresses using UDS/LDS)

I'm working on an input system which should allow blindly typing random stuff and having it spoken.

Anna: you mentioned some problem with recording wavefiles from dectalk? can you explain?

LN


LN, you have a PM. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/11/09 06:45 PM

SVN r5692
Game Boy Color (gbc driver)
Game : Zelda - Oracle Of Seasons

Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/11/09 06:47 PM

LN: AFAIK you can't mask an AM_RAM like that. Aaron?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/11/09 07:01 PM

DECTalk DTC-01 Speech Sound Sample

Thanks LN. smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/11/09 07:11 PM

Originally Posted By Anna Wu


That's so cool!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/11/09 08:02 PM

SVN r5696
Game Boy Color (gbc driver)


Posted By: Shideravan

Re: SVN builds - new driver flood - 09/11/09 10:34 PM

Wow!
Awesome!
You are the bests of the bests!
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 09/11/09 10:46 PM

Be warned that the gameboy color bios we currently have is not the 'real' one: it was ?accidentally? left in the rom of pokemon stadium for n64, possibly a leftover from an older version of that hardware which actually had a varant of the cgb cpu in it; the bios is 'incomplete': it doesn't do the near-constant cart logo or header checks which the real cgb bios does, and it doesn't have the table of various different color palettes predone for games which are selected based on a ?hash? of the header title of each game. It also does NOT result in the correct register states when the cart code is started which may cause all sorts of weird errors.
In fact, I don't even know if the 'pokemon stadium cgb bios' even checks the header flags to set up the 'undocumented' registers to switch the windowing and sprite priority registers to the proper modes for cgb vs dmg games. it probably doesn't.

Based on preliminary decapsulation work on the cgb cpu, the bios is 3 roms:
a 256 byte one (probably from 0x0000-0x00FF) which is similar to the dmg one
a 512 byte one (probably from 0x0150-0x034F) which holds the logo and code to decompress it
and a 1792 byte one (address unknown so far) which holds the big lookup table of games and what colors they should have.
None of these roms have actually been read out yet due to difficulty staining the die, but we know how large they are.

LN
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/12/09 01:06 AM

SVN #5699 says...

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/12/09 05:56 AM

Thanks Arbee. Some games like Pinobee: Wings of Adventure looks now better
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/12/09 06:26 AM

Originally Posted By Anna Wu
SVN r5692
Game Boy Color (gbc driver)
Game : Zelda - Oracle Of Seasons



After compiling r5702, something broken on gbc driver again.
See the Boot-Logo and then a white screen,
With r5692 and r5696*, i was able to load and play games like Zelda - Oracle Of Seasons.

* With r5696 I am not 100% sure, because can not remember me. But I think it was working, otherwise i had post the problem already before.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/12/09 12:43 PM

SVN r5702
PC Engine (pce driver)
Game : The Akumajou Dracula X - Chi no Rondo (CD-ROM/CHD)



How i created the needed CHD file :

Daemon Tools (as example) for CD/DVD emulation (Windows)
burnatonce (as example) to write CD (Windows)
chdman to convert .toc to chd

1.) Start Daemon Tools and load image.
2.) Start burnatonce, choose drive for CD emulation and read to image.
3.) Start chdman to convert the .toc file.
Example: chdman -createcd Dracula_X.toc Dracula_X.chd
4.) Start MESS and choose for cartslot as example: Syscard3.pce and for CDROM as example: Dracula_X.chd
5.) Start the emulation and push the RUN button to start the game.

Unfortunately, the game is not playable.
Posted By: Shideravan

Re: SVN builds - new driver flood - 09/12/09 01:57 PM

Hum!
Thank you...

I tried the daemon tools but don't seens t run in windows seven.
Alchool 120% i don't have the license anymore...

Do you know another virtual drive?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/12/09 01:59 PM

Originally Posted By Shideravan
Hum!
Thank you...

I tried the daemon tools but don't seens t run in windows seven.
Alchool 120% i don't have the license anymore...

Do you know another virtual drive?


Sorry, I use WinXP Pro.
Maybe Gizmo Drive ?
Posted By: Shideravan

Re: SVN builds - new driver flood - 09/12/09 02:02 PM

Ok, I'll search more for that...
Posted By: Kale

Re: SVN builds - new driver flood - 09/12/09 06:01 PM

r5700 [X1] Added very basic .TAP cassette image support.

Thanks to Barry, now the "most famous sleeper game on Sharp X-1" (Kagirinaki Tatakai) is finally working. wink



Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/12/09 06:26 PM

Cool! smile
Posted By: Shideravan

Re: SVN builds - new driver flood - 09/13/09 12:48 AM

SVN r5710
TurboGrafx 16
Game: Air Zonk

Regression:
The game begins normally but in the first stage, after a few seconds, just brokes...

Game: Cadash
Graphics errors...
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/13/09 10:09 AM

either you give us a version where it was working or it's not a regression. in particular I seem to remember both bugs being present since early judge's work on the driver...

I might be wrong however
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/13/09 01:54 PM

Originally Posted By Anna Wu
Originally Posted By Anna Wu
SVN r5692
Game Boy Color (gbc driver)
Game : Zelda - Oracle Of Seasons



After compiling r5702, something broken on gbc driver again.
See the Boot-Logo and then a white screen,
With r5692 and r5696*, i was able to load and play games like Zelda - Oracle Of Seasons.

* With r5696 I am not 100% sure, because can not remember me. But I think it was working, otherwise i had post the problem already before.


judge, please can you take a look ?
Posted By: judge

Re: SVN builds - new driver flood - 09/13/09 06:52 PM

Originally Posted By Anna Wu
Originally Posted By Anna Wu


After compiling r5702, something broken on gbc driver again.
See the Boot-Logo and then a white screen,
With r5692 and r5696*, i was able to load and play games like Zelda - Oracle Of Seasons.

* With r5696 I am not 100% sure, because can not remember me. But I think it was working, otherwise i had post the problem already before.


judge, please can you take a look ?


Just like some other things this seems to hang waiting for some interrupt to occur. Some interrupts occur and get handled but apparently not the correct interrupts. I'm not sure if I can fix this today.
Posted By: Shideravan

Re: SVN builds - new driver flood - 09/14/09 03:53 AM

Originally Posted By Anna Wu

Unfortunately, the game is not playable.


No problems, Anna!
At less the intro animation and the musics is running well!
Very impressive for that time!
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 09/14/09 06:17 AM

R.Belmont: the ESQ-1 driver needs a 2681 Duart: the 68681 duart core will handle this just fine, just set the interrupts up to ignore the 'vector' parameter. Otherwise the 68681 and 2681 duarts are practically completely identical (the only differences are detailed in the appendix A of this datasheet: http://www.cs.indiana.edu/csg/hardware/c335_tools/MC68681UM.pdf ; in short: the 68681 duart is 68k aware, hence has an /INTACK pin and an IVR register, while the 2681 is general purpose and has an IP6 pin instead, and the *reserved* reg in place of the IVR reg doesn't do anything except store a byte of data. Also the 2681 has a RESET pin and 68681 has /RESET pin, and the secondary purposes of the IPx pins are shuffled around a bit)

Be warned that the 68681 core is still missing a number of features: there is no way (that I've found) to push an interrupt state update if it has delta interrupts set up for the ip0/ip1/ip2/ip3 input pins; the hardware rts/cts control for channels A/B based on pins ip0/ip1 and op0/op1 (controlled by MR2A/B bits 4 and 5) is also not implemented yet, but I plan to add this soon.
Also the counter/timer modes which depend on the input pins are not implemented, and implementing them would seem to require a full state machine emulation of the core, which would require a ground-up rewrite. Also emulating actually reading the counter registers will be a little tricky, though timed interrupts do work right now using timers for the full count duration.

LN
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 06:31 AM

SVN r5722
GP2X 2.0 (gp2x skeleton driver)

Something is coming ... smile

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 06:47 AM

SVN r5724
Sharp X1 (x1 driver)
Game : DigDug (Cassette)

The tape is now working if the DIP switch: System/IOSys is on. smile



PS: Is it possible to add tap to wav conversion in castool ?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 10:23 AM

SVN r5727
Compiling error

Compiling src/mess/drivers/vip.c...
cc1.exe: warnings being treated as errors
src/mess/drivers/vip.c: In function 'vip_sc_w':
src/mess/drivers/vip.c:538: warning: unused variable 'driver_sta
mingw32-make: *** [obj/windows/mess/mess/drivers/vip.o] Error 1

Edit : it´s now fixed.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 10:46 AM

SVN r5727
SAM Coupe (samcoupe driver)
Game : Kaboom (Cassette)

Posted By: Duke

Re: SVN builds - new driver flood - 09/14/09 11:03 AM

Did you convert a .tap file to .wav or did you find the image already in .wav format Anna?

I actually only tested saving and loading via the builtin BASIC, nice to see it works for already existing images too smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/14/09 11:07 AM

is there any particular reason for rev. 5725? I think we should have the users pressing PLAY and STOP at the right moment as in the real thing (see c64) and not autoplaying...

then we need, of course, good instructions on the wiki for systems which need these keys pressed wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 11:23 AM

Originally Posted By Duke
Did you convert a .tap file to .wav or did you find the image already in .wav format Anna?

I actually only tested saving and loading via the builtin BASIC, nice to see it works for already existing images too smile


Very easy. I used Tape2WAV v1.8 vor Windows to convert .tap file to .wav.
The conversion for SAM Coupe tap format is missing and the ZX Spectrum conversion for this tap format is not working in/with castool.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 11:39 AM

You have a PM, Duke.
Posted By: Duke

Re: SVN builds - new driver flood - 09/14/09 12:05 PM

Thanks Anna, I have added support for .tap images now.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 12:15 PM

Originally Posted By Duke
Thanks Anna, I have added support for .tap images now.


For driver/emulation or for castool ?

Edit: I see, in the driver. smile
Posted By: JoJo

Re: SVN builds - new driver flood - 09/14/09 12:26 PM

Originally Posted By Anna Wu
Originally Posted By Duke
Thanks Anna, I have added support for .tap images now.


For driver/emulation or for castool ?

Edit: I see, in the driver. smile


cassimg.c includes a cassette_dump function. Currently it depend on a #define DUMP_CASSETTES. Why don't transform it into a full-fledged debugger command?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/14/09 06:41 PM

SVN r5734
PC-8801 emulation.
Someone can boot floppy disks ?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/16/09 04:43 PM

Originally Posted By Anna Wu


Anna, I do not mean to sound ungrateful, but is it possible that you could find a new file host for audio clips that does not pop up several windows claiming that people have a virus and need to scan their computers, and then try to infect people with spyware, even on Firefox, when trying to download that file?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/09 05:04 PM

No problem, for the next sound file I will find and use a alternative.
Posted By: judge

Re: SVN builds - new driver flood - 09/16/09 05:09 PM

Originally Posted By judge
Originally Posted By Anna Wu
Originally Posted By Anna Wu


After compiling r5702, something broken on gbc driver again.
See the Boot-Logo and then a white screen,
With r5692 and r5696*, i was able to load and play games like Zelda - Oracle Of Seasons.

* With r5696 I am not 100% sure, because can not remember me. But I think it was working, otherwise i had post the problem already before.


judge, please can you take a look ?


Just like some other things this seems to hang waiting for some interrupt to occur. Some interrupts occur and get handled but apparently not the correct interrupts. I'm not sure if I can fix this today.


Fixed the gameboy color driver. The hacked up bad dump of the bios was confusing most color games.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/16/09 07:26 PM

SVN r5745
Game Boy Color (gbcolor driver)
Games : Zelda - Oracle Of Ages, Zelda - Orcale Of Seasons,
Super Mario Bros. Deluxe

Great work, judge ! laugh


Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/17/09 11:30 AM

SVN r5763
Super Game Boy (supergb driver)
Game : Warioland II (running with new Super Game Boy boot rom)

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/17/09 02:18 PM

SVN r5766
Apple IIgs (ROM01) (apple2gsr1 driver)
Game : Arkanoid (Floppy Port 3 (Sony))



Apple IIgs (ROM03) (apple2gs driver)
Game : Bard's Tale II (Floppy Port 3/4 (Sony))

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/17/09 05:27 PM

SVN r5770
SNES driver
Games : Mega Man X 2 (U), Mega Man X 3 (U) with CX4 chip.


Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/17/09 05:40 PM

Don't get your hopes up quite yet, they aren't really playable, as any wireframe stuff just shows up as garbage. I'll look at it this weekend. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/17/09 05:44 PM

Originally Posted By Just Desserts
Don't get your hopes up quite yet, they aren't really playable, as any wireframe stuff just shows up as garbage. I'll look at it this weekend. smile


Just finished the end boss. smile

Seriously, it is a good start.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/17/09 05:59 PM

Originally Posted By Anna Wu
Originally Posted By Just Desserts
Don't get your hopes up quite yet, they aren't really playable, as any wireframe stuff just shows up as garbage. I'll look at it this weekend. smile


Just finished the end boss. smile

Seriously, it is a good start.


Wow, you're good at Mega Man. laugh
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/17/09 06:45 PM

It was a joke. wink
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/18/09 12:05 AM

SVN #5780



Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/18/09 12:10 AM

Ooh, aah smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/18/09 01:06 AM

I can't get any of my floppy disk images to boot on the apple2gs driver now. Is there some trick?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/09 06:16 AM

Originally Posted By R. Belmont
I can't get any of my floppy disk images to boot on the apple2gs driver now. Is there some trick?


The apple2gs driver use 2 different floppy disk drive types (Apple and Sony)
My games was running on floppy disk devices 3 and 4 (Sony)
If some disk images are no booting on floppy disk devices 1 and 2 (Apple), try 3 and 4 (Sony).

Example for Arkanoid.dsk :

1st Screenshot: Floppy Disk Device 1 (Apple)
2nd Screenshot: Floppy Disk Device 3 (Sony)



Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/09 06:18 AM

Looks nice, Just Desserts. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/09 09:11 AM

Originally Posted By Anna Wu
SVN r5763
Super Game Boy (supergb driver)
Game : Warioland II (running with new Super Game Boy boot rom)



SVN r5784

It seems, some problems with new Super Game Boy boot rom exist.

Posted By: judge

Re: SVN builds - new driver flood - 09/18/09 09:12 AM

Yes, that was also mentioned in the commit message for that revision. I'll be taking a look later today.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/09 09:13 AM

Originally Posted By judge
Yes, that was also mentioned in the commit message for that revision. I'll be taking a look later today.


Thanks for your help. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/18/09 01:41 PM

Originally Posted By Anna Wu
The apple2gs driver use 2 different floppy disk drive types (Apple and Sony)
My games was running on floppy disk devices 3 and 4 (Sony)
If some disk images are no booting on floppy disk devices 1 and 2 (Apple), try 3 and 4 (Sony).


Oh, I see. flop1/2 are s6d1/2 and flop3/4 are s5d1/2. The old naming made waaaay more sense to those of us who've used real hardware smile

Posted By: Justin

Re: SVN builds - new driver flood - 09/18/09 02:43 PM

The naming probably just got overlooked in the floppy overhauls.
Posted By: Micko

Re: SVN builds - new driver flood - 09/18/09 02:57 PM

Robert : I am aware of name change, but thing is that my idea was that in future, you have logic name of floppy in floppy config.

Also idea is to set size (3", 3.5", 5.25", 8",...) and supported density (SD,DD,HD,ED) and head number (1,2).

That way I can put some DSSD disk in SSSD drive but can read one side only (will show warning in this kind of situation).
But can't put 3.5" disk in 5.25 drive (some formats contain this info).

So lot of changes will occure, and I will try to keep as much as possible old info inside.
Posted By: judge

Re: SVN builds - new driver flood - 09/18/09 06:37 PM

Originally Posted By Anna Wu


SVN r5784

It seems, some problems with new Super Game Boy boot rom exist.


The boot screen is fixed, but somehow it doesnt display in color.
Posted By: judge

Re: SVN builds - new driver flood - 09/18/09 07:27 PM

supergb is displaying super gb games in color again.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/09 09:06 PM

SVN r5794
Super Game Boy (supergb driver)
Games : Donkey Kong Land (running with new Super Game Boy boot rom)

Great fix, judge.
Thanks also goes to Curt.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/09 09:16 PM

SVN r5779
snes driver
Game : SD Gundam GX (J) with DSP3 chip

Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/18/09 11:12 PM

A challenger appears!

SVN r5797
SNES driver
Game : Star Ocean (J) with S-DD1 chip





Game : Street Fighter Alpha 2 (U) with S-DD1 chip







Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/18/09 11:16 PM

Wow, thank you so much. smile

I lost the overview, which special chip support is now still missing ?

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 09:55 AM

SNES(FX) driver
Games with special chips (not complete) Part 1

Examples:

Mega Man X 2 (U) with CX4 chip



Super Mario Kart (U)with DSP1 chip



Dungeon Master (U)with DSP2 chip



SD Gundam GX (J) with DSP3 chip



Top Gear 3000 (U) with DSP4 chip



Metal Combat - Falcon's Revenge (U) with OBC-1 chip



Games with with SA-1 chip seems not to work.

Street Fighter Alpha 2 (U) with S-DD1 chip



Games with with SPC7110 chip seems not to work.

Dai Kaiju Monogatari 2 (J) with S-RTC chip (partially working)



F1 ROC II - Race of Champions (U) with ST010 chip (partially working)



Hayazashi Nidan Morita Shougi (J) with ST011 chip (partially working)



Games with with ST018 chip seems not to work.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 09:56 AM

SNES(FX) driver
Games with special chips (not complete) Part 2

Examples:


Star Fox (U) with Super FX chip



Super Mario - Yoshi Island (J) with Super FX 2 chip



If some special chips are missing, please tell me.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/19/09 09:58 AM

SA-1, RTC, SPC7110, ST-01x are not supported

only SuperFX, S-DD1, CX4, DSP-x & OBC-1 are expected to work
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 10:35 AM

Originally Posted By etabeta78
SA-1, RTC, SPC7110, ST-01x are not supported

only SuperFX, S-DD1, CX4, DSP-x & OBC-1 are expected to work


Thanks, corrected some.
About Dai Kaiju Monogatari 2 (J) with S-RTC chip, I am not sure.




Posted By: etabeta78

Re: SVN builds - new driver flood - 09/19/09 10:40 AM

the game works, but it does not use currently the S-RTC features (which is a RealTimeClock) which should keep your game synch'd with the current time
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 10:43 AM

Originally Posted By etabeta78
the game works, but it does not use currently the S-RTC features (which is a RealTimeClock)


Changed to

Dai Kaiju Monogatari 2 (J) with S-RTC chip (partially working)
Posted By: Kale

Re: SVN builds - new driver flood - 09/19/09 03:11 PM


r5800 /src/mess/drivers/x1.c: [X1]: Rewritten bitmap routine in order to support tile height feature, this fixes gfxs in Xevious and King's Knight


Xevious




King's Knight

http://mamedev.emulab.it/kale/fast/files/0124.png

Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/19/09 03:23 PM

Awesome smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 08:17 PM

Good news, Kale. smile

PS: Can you take a look to the game " 1942 " (Floppy Disk), what´s happen ?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 08:32 PM

SVN r5802
Compiling error

Compiling src/mess/formats/imd_dsk.c...
cc1.exe: warnings being treated as errors
src/mess/formats/imd_dsk.c: In function 'imd_dsk_identify':
src/mess/formats/imd_dsk.c:39: warning: 'and' of mutually exclusive equal-tests
is always 0
mingw32-make: *** [obj/windows/mess/mess/formats/imd_dsk.o] Error 1
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/19/09 08:47 PM

Micko just checked in a fix.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 08:51 PM

SVN 5804
Compiling Error

Compiling src/mess/formats/imd_dsk.c...
cc1.exe: warnings being treated as errors
src/mess/formats/imd_dsk.c: In function 'imd_read_sector':
src/mess/formats/imd_dsk.c:80: warning: 'offs' is used uninitialized in this function

src/mess/formats/imd_dsk.c:64: note: 'offs' was declared here
mingw32-make: *** [obj/windows/mess/mess/formats/imd_dsk.o] Error 1
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/19/09 09:00 PM

r5806 should contain a fix for that new one. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 09:02 PM

Originally Posted By Just Desserts
r5806 should contain a fix for that new one. smile


Yes, now the problem is solverd. Thanks !
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/19/09 09:21 PM

SVN r5808
SNES driver
Game : Momotarou Dentetsu Happy (J) with SPC7110 chip

Boot Procedure:
1. Press the A button when prompted.
2. Wait.
3. Press F3 when prompted to power off.
4. Press the B button when prompted.
5. Wait.
6. Press F3 when prompted to power off.
7. The game will boot.









I'll let etabeta post the nice pictures of F1 ROC II. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/19/09 09:37 PM

maybe tomorrow. the whole snes driver crawls on my eeepc.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/19/09 09:46 PM

For what it's worth, ST010 doesn't fully work yet. Once you actually get in a race against the CPU, all of the CPU-controlled cars turn fully left and drive off the course. frown
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 09:56 PM

Quote:
SVN r5808
SNES driver
Game : Momotarou Dentetsu Happy (J) with SPC7110 chip

Boot Procedure:
1. Press the A button when prompted.
2. Wait.
3. Press F3 when prompted to power off.
4. Press the B button when prompted.
5. Wait.
6. Press F3 when prompted to power off.
7. The game will boot.


If I press F3, no reaction.

Quote:
I'll let etabeta post the nice pictures of F1 ROC II.


If I start the emulation, MESS freeze (Initializing ...)

Posted By: etabeta78

Re: SVN builds - new driver flood - 09/19/09 10:04 PM

there was at least one typo in the code I sent you, sorry...

Code:
o1 = (UINT8)snes_st010.o1;


should be

Code:
o1 = (UINT16)snes_st010.o1;


this hopefully fixes the CPU AI
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/19/09 10:15 PM

Freeze problem with r5809 and mounted F1 ROC II cart solved.



Still problem with no working F3 key (on my Notebook) and Momotarou Dentetsu Happy game.

Posted By: etabeta78

Re: SVN builds - new driver flood - 09/19/09 10:16 PM

Originally Posted By Anna Wu
Quote:
I'll let etabeta post the nice pictures of F1 ROC II.


If I start the emulation, MESS freeze (Initializing ...)



you need at least rev 5809 because apparently the spc7110 init was conflicting with the st010 one...

EDIT: ok I see you fixed smile
Posted By: Justin

Re: SVN builds - new driver flood - 09/19/09 10:32 PM

src\mame\machine/snes7110.c(517) : warning C4333: '>>' : right shift by too large amount, data loss
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/19/09 11:17 PM

fixed in rev 5811. and what compiler is that?
Posted By: Justin

Re: SVN builds - new driver flood - 09/19/09 11:19 PM

Microsoft Visual C++ 2008 Express Edition
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/19/09 11:24 PM

Ahh, yes. Aaron works at Microsoft and still uses 2005, so there's probably something wrong with 2008 nobody's telling us smile
Posted By: Kale

Re: SVN builds - new driver flood - 09/20/09 01:08 PM

Originally Posted By Anna Wu
Good news, Kale. smile

PS: Can you take a look to the game " 1942 " (Floppy Disk), what´s happen ?


PCG addressing uses a complex system with the beam positions and the videoram bank...there are some games (Bastard SP, Herzog, 1942 and others) that relies on this...
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/20/09 02:03 PM

Can someone familiar with issues involving PC keyboards and MESS please help Anna? For some reason, on her German-layout keyboard, F3 doesn't function as reset, and neither do any of the keys on the keyboard, whether pressed normally, with Shift, with Alt, or both.

I would normally guess that it's a problem with inadvertantly remapping the button, but I searched high and low in the Input menus, and apparently F3 is hard-coded as "Reset", and is not remappable.

Does anyone have any ideas?
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/20/09 02:24 PM

it is remappable (I needed it on my Macbook and used to remap it to COMMA or PERIOD wink )

TAB > Input general > User Interface > Reset Game
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/20/09 02:26 PM

Originally Posted By etabeta78
it is remappable (I needed it on my Macbook and used to remap it to COMMA or PERIOD wink )

TAB > Input general > User Interface > Reset Game


Weird, I didn't see it when I was in there. I stand corrected. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 03:26 PM

Not easy, because I use the New UI/MESSUI.
To get the TAP> Input (General) I have to use the standard GUI.
With the standard UI I can map the keyboard key F3.
Unfortunately, the mapped key I cannot use in New UI/MESSUI.

Anyway, thanks for your great help. smile



Posted By: etabeta78

Re: SVN builds - new driver flood - 09/20/09 03:29 PM

Sooner or later, New UI will be removed. you might consider starting to get used to Standard MAME UI with MESSUI wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 03:32 PM

@Michael

Last SVN
ti99_4a driver

It seems, something is broken with the Floppy Disk support.
Example:
I am no more able to boot UCSD p-System disks. P-Code card is on.
Will check also other disks (Assembler/Basic) and ti99_xx driver.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 03:34 PM

Originally Posted By etabeta78
Sooner or later, New UI will be removed. you might consider starting to get used to Standard MAME UI with MESSUI wink


I hope later, much later.
Posted By: Shideravan

Re: SVN builds - new driver flood - 09/20/09 03:43 PM

The old UI is much better!
But i think that could be a way from write the name in cart selection on file manager, like the MAME/MESS without UI...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 03:54 PM

Originally Posted By Anna Wu
@Michael

Last SVN
ti99_4a driver

It seems, something is broken with the Floppy Disk support.
Example:
I am no more able to boot UCSD p-System disks. P-Code card is on.
Will check also other disks (Assembler/Basic) and ti99_xx driver.


OK, it seems only boot disk problem exist, if the P-Code card is activated.
Without the P-Code card, I can boot disks.

The problem not exist with the official MESS 0.134 release.
Posted By: Micko

Re: SVN builds - new driver flood - 09/20/09 03:57 PM

This could be result of my recent changes to device.
Just please send me needed files on email, so I could fix.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 04:01 PM

Originally Posted By Micko
This could be result of my recent changes to device.
Just please send me needed files on email, so I could fix.



Check your mail.
Posted By: mizapf

Re: SVN builds - new driver flood - 09/20/09 04:27 PM

Oh, is this my dear old friend "double-steps" that you resurrected?

Seems so as all my disks fail again after reading halfway. I have to look at the code.

Michael

P.S.: Argh ... what happened to mflopimg.c? This had the vital part in it, to make the floppy work again for the ti99 family.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 04:51 PM

Quote:
Seems so as all my disks fail again after reading halfway. I have to look at the code.


I can boot my disks by using the E.Basic/Assembler cartridge.
or you are talking about PCode-Card disks ?

Posted By: Micko

Re: SVN builds - new driver flood - 09/20/09 04:51 PM

mizapf: All is still there translate function is set as before.

Anna:
Strange thing is I can not activate pCard at all, I set it in dip switch but when I look again it is off.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 05:00 PM

Originally Posted By Anna Wu
Quote:
Seems so as all my disks fail again after reading halfway. I have to look at the code.


I can boot my disks by using the E.Basic/Assembler cartridge.
or you are talking about PCode-Card disks ?



I tried more disks.
You are right, some disks i can boot and some can not as example TI-Artist. So problem not only for PCode-Card exist.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 05:04 PM

Originally Posted By Micko
mizapf: All is still there translate function is set as before.

Anna:
Strange thing is I can not activate pCard at all, I set it in dip switch but when I look again it is off.


Maybe this thread can be helpful. Anyway, I will send you the .cfg file.
Posted By: mizapf

Re: SVN builds - new driver flood - 09/20/09 05:39 PM

You can verify the problem when using the Disk Manager cartridge and do a disk check. It will fail after half of the tracks.

@Micko: It is not a matter of the track translate function - the problem is that the TI disk controllers try to handle drive/media track count mismatches. Specifically, if the drive can handle 80 tracks and you insert a 40 track disk, the controller ROM tries to detect that and will apply double physical steps for each logical track step. This is a behaviour from the TI ROMs, not from the MESS emulator.

Version 0.98 MESS introduced code on the assumption that one can safely set the media track count equal to the drive track count - which instantly broke the whole TI emulation (until v0.123 where I fixed it). My solution was to include a check in mflopimg.c, internal_floppy_device_load (look for DEVINFO_INT_KEEP_DRIVE_GEOMETRY). Have you moved that somewhere else?

I am aware that a much better solution would be to create a specific drive/image implementation instead of these tricks using basicdsk, and I already planned to do some work there, but for the time being, please make sure that the drive emulation and the medium emulation allow for different track counts. This is simply mirroring the real situation.

Michael
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 06:19 PM

Quote:
You can verify the problem when using the Disk Manager cartridge and do a disk check. It will fail after half of the tracks.


SVN r5817
ti99_4ae driver
Pitfall (Floppy Disk)
Disk Manager 3 (Cartridge)

I can boot the disk and play the game
but

if I check the tracks with the Disk Manager, I get errors.

1st Screenshot: SVN r5817, 2nd Screenshot: official MESS 0.134


Posted By: Micko

Re: SVN builds - new driver flood - 09/20/09 06:24 PM

Michael:

You are right, it is exacly that option that is missing, looking now for easiest way to add it, but it seams I need to update all floppy defs for drivers, with new param.
Posted By: Micko

Re: SVN builds - new driver flood - 09/20/09 06:41 PM

Should be ok now. Parameter added to all configs.
Posted By: mizapf

Re: SVN builds - new driver flood - 09/20/09 07:15 PM

@Micko: Apologies in the name of the TI systems for causing these troubles. As I said, I already planned to clean up the handling, but that turned out not to be so easy as I had to dig deeply into the call stacks and to learn a lot about the device handling in MESS. I think I found a call with depth 20 for reading a sector somewhere, so that's not really something for the coffee break.

I think I have now understood the major part, will try to catch up with your changes. But I would appreciate if you can achieve some workaround for now, promising that I will take care of the restructuring afterwards.

Some more details which I found out two years ago when working on the fix: The TI system allows for industry floppy drives, no special devices from TI, just the controllers. Earlier controllers (with wd17xx) only allow 40 tracks, later controllers (smc92x4) also allowed for 80 track drives. Instead of doing a test seek, the creator of the advanced controller decided to make use of dip switches to inform the controller about the number of physical tracks of the attached drive.

We now have a majority of disk images from 40 track drives and also some images from 80 track drives. Consequently, we need to support the 80 track drives which must also cope with 40 track images.

The disk controller reads the actual number of tracks from an entry in sector 0 of the disk. If it finds it as 40 tracks (from the disk image) but it is an 80 track drive (from the dip switch), it knows that it will have to issue two seek actions per medium track. This is done by the disk controller ROM.

If you now override the device track count with the medium track count, you will get into trouble: From the 40th track on, the drive will refuse any further seek commands, but the controller believes to be at medium track 20. This will raise an error condition. As SSSD disk images on the TI have 360 sectors, you get a seek error from sector 180 upwards.

Now for Anna's observation: The disk directory is located in early sectors, so you have a good chance that the directory is fully intact, and so are all files below the middle of the disk. Within the image emulation, the track translation maps the track count back to the logical track number/2, so everything is again readable up to that point. But any file that extends beyond the middle track will be unreadable.

Michael
Posted By: mizapf

Re: SVN builds - new driver flood - 09/20/09 07:20 PM

@Micko: Confirm. Disk tests show 0 errors now.
Good work!

Michael
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/20/09 07:23 PM

Also for me, everything is working now.

Let us change your status, Micko.
From doctor to chief doctor. smile
Posted By: JoJo

Re: SVN builds - new driver flood - 09/20/09 11:02 PM

Originally Posted By mizapf


Some more details which I found out two years ago when working on the fix: The TI system allows for industry floppy drives, no special devices from TI, just the controllers. Earlier controllers (with wd17xx) only allow 40 tracks, later controllers (smc92x4) also allowed for 80 track drives. Instead of doing a test seek, the creator of the advanced controller decided to make use of dip switches to inform the controller about the number of physical tracks of the attached drive.

Michael


This observation suggests that the TI99 drives should be devices with their own PORT_DIPSETTING entry: is this something possible with the actual core? Do the "Dipswitches" and "Configuration" menu allow sub-menus?
Posted By: Kale

Re: SVN builds - new driver flood - 09/20/09 11:02 PM


r5823 /src/mess/drivers/x1.c: [X1TURBO]: Implemented 2 bpp colors on tilemap layer


Gradius (x1turbo version), before...


...and after


Dunno how many games uses this feature, probably not many of them...
Posted By: mizapf

Re: SVN builds - new driver flood - 09/21/09 07:06 PM

Originally Posted By JoJo
This observation suggests that the TI99 drives should be devices with their own PORT_DIPSETTING entry: is this something possible with the actual core? Do the "Dipswitches" and "Configuration" menu allow sub-menus?

that's something which also came to my mind. Currently I'm using dip switches for selecting emulated devices (which in reality means attaching/detaching the device), and also for configuring, representing real dip switches. Maybe there is potential for more structure in configuration, like grouping dip switches with devices, only offering them if the device is actually attached.

Michael
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/21/09 07:21 PM

one thing you can do is to use a PORT_CONFIG to attach the device and then add PORT_CONDITION to the settings which depend from this (see e.g. coco.c or many MAME drivers for the use of PORT_CONDITION)

also, ideally, you should use Dipswitches only for physical dips, the other settings should use Configs (not sure if you already 'follow' this convention with ti99)
Posted By: Braille

Re: SVN builds - new driver flood - 09/21/09 09:45 PM

About r5838, .sg files are already used in SG-1000.

File sg1000.c has: MDRV_CARTSLOT_EXTENSION_LIST("sg,bin")

The revision changed Master System file (sms.c). Guess something's wrong.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/21/09 10:08 PM

it's just that in theory the japanese sms, which was called SG1000 Mark III should be backward compatible with games for SG1000 and SG1000 Mark II, i.e. .sg files

not sure how compatible emulation is currently
Posted By: Braille

Re: SVN builds - new driver flood - 09/21/09 10:28 PM

Hum, I see. Although SG1000 Mark III already opened .sg files (and their .zip), there wasn't an option to do so in default. Thanks!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/22/09 04:18 PM

SVN r5850
Compiling error :

Linking messlib.dll...
obj/windows/mess/mess/shared.a(flopdrv.o):flopdrv.c:(.text+0x339): undefined ref
erence to `mess_devclass_from_core_device'
collect2: ld returned 1 exit status
mingw32-make: *** [messlib.dll] Error 1



Posted By: Micko

Re: SVN builds - new driver flood - 09/22/09 04:44 PM

Anna:

Just try clean build. There is no error
You can always check at :

http://mess.redump.net/mess:compile_failure
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/22/09 05:48 PM

Originally Posted By Micko
Anna:

Just try clean build. There is no error
You can always check at :

http://mess.redump.net/mess:compile_failure


OK, i make a clean build.
Will answer you tomorrow, because to make a clean build need some hours on my slow Notebook. frown
Posted By: judge

Re: SVN builds - new driver flood - 09/22/09 05:51 PM

Try adding a -j3 to the make command; that may speed up the build a bit
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/22/09 06:56 PM

Fair warning: -j3 is for dual cores (or hyperthreading). Don't do it on a single-processor system, it'll make things slower smile
Posted By: judge

Re: SVN builds - new driver flood - 09/22/09 07:00 PM

Are there still non-dual core machines then? wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/22/09 08:27 PM

Unfortunately I have an Intel Mobile Pentium 4 processor with 1.6 GHz.
Compiling is still in progress ...

Edit: Not use the parameter : -j3
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/23/09 07:45 AM

Now I have a clean build. smile
Was not easy, because I deleted only the bin\mess\obj\ folder before.
But is was not enough, so i deleted also the Bin\mess\src\mess\formats\ folder and some files in bin\mess\src\mess\drivers\ folder.

Quote:
H:\SVN\Bin\mess>mingw32-make TARGET=mess
mingw32-make: Nothing to be done for `all'.


Posted By: judge

Re: SVN builds - new driver flood - 09/23/09 07:48 AM

You should let the makefile clean things for you:
Code:
mingw32-make TARGET=mess clean


That will clean all the generated binaries and object files.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/23/09 07:52 AM

Originally Posted By judge
You should let the makefile clean things for you:
Code:
mingw32-make TARGET=mess clean


That will clean all the generated binaries and object files.


Oh yes, now I remember me. It is much easier as do it manually.
Next time ... smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/23/09 08:02 AM

however deleting the whole obj/ folder and the executable is more than enough (I think it is exactly what the "clean" do).

deleting part of the source is not needed because the message

mingw32-make: Nothing to be done for `all'.

appears when the compiler founds all the files in the obj/ directory and no newer source file. When you delete obj/, the compiler will compile the whole source again because it finds nothing in the obj/ folder
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/23/09 08:35 AM

Originally Posted By etabeta78
however deleting the whole obj/ folder and the executable is more than enough (I think it is exactly what the "clean" do).

deleting part of the source is not needed because the message

mingw32-make: Nothing to be done for `all'.

appears when the compiler founds all the files in the obj/ directory and no newer source file. When you delete obj/, the compiler will compile the whole source again because it finds nothing in the obj/ folder


This message was coming, after i delete some source files too. smile
Before, I can compile but if I compile as example the 2nd time, always the same files was compiled again.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/23/09 09:12 AM

once the source files are compiled, they generate .o files in the obj directory. after that, they are not compiled anymore until you remove that .o file or update the source files (removing files does not count as update [1]). that's why it is the obj/ directory that must be changed


[1] on the other hand opening, changing something, undoing the change and saving the file would force the compiler to recompile the file even if it is the same as before, because the file has been saved since last time it's been compiled
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/23/09 05:51 PM

SVN r5865

Epson PX-4 driver
Basic Cartridge



Epson PX-4+ driver
PX4Util Cartridge


Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/24/09 12:47 PM

SVN r5877
800 Junior (elwro800) driver

Spectrum mode
Ms. Pacman (Floppy Disk)



CP/J mode (similar to CP/M)
Vulcan Calc (Floppy Disk)



Posted By: Stiletto

Re: SVN builds - new driver flood - 09/24/09 01:01 PM

Originally Posted By Anna Wu
SVN r5865

Epson PX-4 driver
Basic Cartridge



Epson PX-4+ driver
PX4Util Cartridge




I love MESS's LCD simulation. I forget, can they optionally output black on white? Do all handhelds that have LCDs have the simulation available?

- Stiletto
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/24/09 01:15 PM

Not know.
Maybe a inverse screen mode not exist.
Posted By: Stiletto

Re: SVN builds - new driver flood - 09/24/09 01:28 PM

I see it. I figured it was something like that.
http://git.redump.net/cgit.cgi/mess/tree/src/mess/layout/px4.lay
Edit values as desired and recompile.

I need to ask Mr. Do about how layouts take precedence over each other. Assuming they can, you could place an "INVERT" layout per LCD system in the layout folder, or a layout which makes the LCD per monochrome black-on-white. If they can't, then you remove the precompiled layout from the source and add one after the fact.

Still curious about whether all monochrome/greyscale LCD handheld systems have layouts that determine the coloring, or if some still need made.

- Stiletto
Posted By: JoJo

Re: SVN builds - new driver flood - 09/24/09 01:36 PM

Originally Posted By Stiletto
Originally Posted By Anna Wu
SVN r5865

Epson PX-4 driver
Basic Cartridge



Epson PX-4+ driver
PX4Util Cartridge




I love MESS's LCD simulation. I forget, can they optionally output black on white? Do all handhelds that have LCDs have the simulation available?

- Stiletto


It's just a matter of enforcing square pixels (there's a layout for this) and chose two colors that mimic as closely as possible the original colors of the LCD - the HP48 drivers go in a lengthier effort to emulate the 16 shades of their peculiar display.

The CoCo3 driver implements the layout concept to emulate the palette switch that occurs when the machine is connected to a RGB or a Composite monitor - maybe something similar could be done in order to switch between B&W, W&B and photorelistic palettes on LCD drivers-

I might be wrong, but I guess that such a patch would be rejected since this trick is not strictly connected to the emulation of the real system...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/24/09 01:38 PM

SVN r5879
Bashkiria-2M (b2m) driver
CP/M 1.12 (Floppy Disk)
Game : Sokoban



Posted By: judge

Re: SVN builds - new driver flood - 09/24/09 08:37 PM

Some more wip screens; special palette selection by the gbc bios for certain mono games:



Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/24/09 09:37 PM

Cool. How was the BIOS dumped ultimately - clock glitching, optical readout, something else entirely? smile
Posted By: judge

Re: SVN builds - new driver flood - 09/24/09 09:38 PM

clock & vcc glitching and a bit of luck
Posted By: Justin

Re: SVN builds - new driver flood - 09/25/09 05:50 AM

That layout doesn't affect the colours for the main display, that stuff is hard-coded into the palette of each driver. I've been wondering whether it might be more appropriate for drivers like e.g. Game Boy to just output greyscale and have a colour overlay in the layout file to fake the LCD shades, but I haven't gotten around to actually trying it to see if it's doable or not.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/09 09:10 AM

CDTV driver
All MESS versions after version 0.129

I have a problem to load successful CD images (CHD).
If I load a image, I get a crash.
For each version, I used always the related version of chdman to create a chd image.

Example:
chdman -createcd welcome.toc welcome.chd

With the official version 0.129 I have no any problems.



Can someone confirm the problem ?



Posted By: box

Re: SVN builds - new driver flood - 09/25/09 11:08 AM

CDTV issue confirmed with latest SVN, works without an image, but MESS chrashes if you provide a CD (chd).
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/25/09 01:03 PM

Originally Posted By box
CDTV issue confirmed with latest SVN, works without an image, but MESS chrashes if you provide a CD (chd).


Thanks !
Posted By: Robbbert

Re: SVN builds - new driver flood - 09/27/09 01:29 PM

At last, Microbee can boot CP/M

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/27/09 06:51 PM

SVN r5901
PC6001 driver

@Kale

It is great, you are working on the PC6001 driver. I like some games for this system. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/28/09 11:48 AM

SVN r5890+
gbcolor driver
Game : Legend of Zelda (.gb) and Super Mario Land (.gb)


Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/28/09 08:23 PM

SVN r5890+
gbcolor driver

Here different selectable color palettes for Legend of Zelda (.gb).



Posted By: judge

Re: SVN builds - new driver flood - 09/28/09 08:32 PM

hehe, cool, anna wu smile
Posted By: ht1848

Re: SVN builds - new driver flood - 09/28/09 08:38 PM

yeah I dig the all the screen shots you post Anna. Thanks!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 09/28/09 09:10 PM

BTW, Trilobit has a new 2600 intro out. Seems to work fine on the a2600p driver. It was intended to be the menu for a demo collection cart, but on real h/w the 2600 bank-switch carts come up in an undefined state (I guess there's no RESET line to the cart?) so they couldn't make it work.
Posted By: JoJo

Re: SVN builds - new driver flood - 09/28/09 09:33 PM

Anna Wu is the new Andy Warhol! wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/29/09 09:56 AM

mz800/mz1500 driver

If I start the emulation, I get following message:
Physical width 920, height 738, nothing more.

Is it possible to solve this problem ?
Posted By: Robbbert

Re: SVN builds - new driver flood - 09/29/09 10:19 AM

turn off error log to get the real message. That physical width thing is a bug in mame core that only shows up in messui (and maybe in mameui but i never use it).


MZ800 gives me a screen full of graphic garbage
MZ1500 gives a black screen.

They are marked as not working, so all is as expected.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/29/09 12:07 PM

Thanks, now i have no more this message. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/29/09 12:08 PM

SVN r5912

MZ700 driver
Cassette
Game : Bomberman (.M12)



Cassette
Game : Mac Pac (.MZF)



Cassette
Game : Serpents (.WAV)



Unfortunately, some good games like Xevious, Space Harrier or Side Roll-F are not working.

MZ80A driver
Cassette
Game : Micropede (.WAV)



MZ80K driver
Cassette
Game : Sargon Chess (.WAV)

Posted By: Anna Wu

Re: SVN builds - new driver flood - 09/29/09 12:35 PM

Sorry, Side Roll-F is working. I will check some other games again.

MZ700 driver
Cassette
Game : Side Roll-F (.M12)



Edit: Xevious and Space Harrier are not working.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 09/29/09 06:53 PM

With latest in SVN: smirk

(click each for big)









Now I just need to figure out or need help figuring out how it actually polls for discs...
Posted By: Kale

Re: SVN builds - new driver flood - 09/29/09 10:35 PM


r5918 /src/mess/drivers/pc6001.c: [PC6001]: Rewritten the video emulation from scratch, ASCII's AX-1 Demo now displays correct gfxs








PS: any help / info / suggestion about how to handle these ADC'ed tape images in MESS is appreciated.
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/29/09 10:52 PM

is there any info on the specific format?

usually, we convert back tape formats to wav at loading time (not always easy, but then you can use core functions to handle everything else wink ), but of course you need to have enough data to recover the original wave (or a square wave that a PC6001 could not distinguish from a real analogic one)
Posted By: Kale

Re: SVN builds - new driver flood - 09/29/09 11:02 PM

pc-6001 .cas format is just raw image in digital format that can be hooked up like a ROM sequencer in MESS (like is currently doing right now).
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/29/09 11:08 PM

yet, given the bad experiences with some old drivers like cgenie or trs80 (and commodore tapes as well), I think it would be better to try converting the bits back to a wave then to feeding bits directly in the memory...

I'll take a look to the format once I'm done with NES mappers, later this week (of course, if you haven't already solved the issue)
Posted By: Kale

Re: SVN builds - new driver flood - 09/30/09 12:33 AM

r5920 /src/mess/drivers/pc6001.c: [PC6001]: Implemented 128x192x2 support, most games uses this mode





They are actually static images, they hangs due of a missing irq (vector 0x06), I should study what this irq does and implement it...
Posted By: etabeta78

Re: SVN builds - new driver flood - 09/30/09 06:14 AM

cool!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/01/09 06:42 PM

SVN r5934
PC6001 driver
Game : Magnetic Field/Raita (Cassette)

Thank you Kale, to explain me how i can add a static image. smile


Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/02/09 05:33 AM

SVN r5938
fmtowns driver (still in progress)



Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/02/09 12:08 PM

Since the version 0.133 I have problems to load harddisk images (.chd) on pc or compatible drivers like pcmda.
The .chd files are already updated with chdman.
As I said, the last working version was 0.132
Posted By: Kale

Re: SVN builds - new driver flood - 10/02/09 06:47 PM

r5942 /src/mess/drivers/pc6001.c: [PC6001]: Implemented cartslot support


Canyon Climber



I'll eventually need this for the exbasic cart image on later models.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/03/09 12:19 AM

Nice. You'd go blind playing that game though smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/03/09 08:45 AM

Originally Posted By Anna Wu
SVN r5938
fmtowns driver (still in progress)





SVN r5949 respectively r5951
fmtowns driver (still in progress)
Color palette fixed.



UNZ 0.5 L10a emulator

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/03/09 05:09 PM

SVN r5952

atvga driver (still in progress)
Game : Atomix (Floppy Disk)



xtvga driver (still in progress)
Game : Prince of Persia (Floppy Disk)

Posted By: Shideravan

Re: SVN builds - new driver flood - 10/03/09 07:12 PM

Wow!
Amazing!
Whats i have to do to run PC/AT?
How I create a HD, etc?
Can i simple grab the software to a created floppy in imgtool to install these things?
Posted By: Justin

Re: SVN builds - new driver flood - 10/04/09 05:34 AM

Hmm, I can't get atvga to boot from disk, I just get lots of horrible beeping and error messages, even if I fool with the keyboard dip switches.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/09 07:02 AM

Originally Posted By Justin
Hmm, I can't get atvga to boot from disk, I just get lots of horrible beeping and error messages, even if I fool with the keyboard dip switches.


It´s a bit tricky, because some problems with the keyboard input. Maybe a interrupt (IRQ) problem but I am not a developer.
I will tell you what I make :
Start the emulation.

Press F1

Wait a while and press N
Now wait a while again, than you will see the cursor is blinking for a very short time faster (Interrupt/IRC ?)
Press than immediately the Enter-Key
If the disk is not booting, soft reset and try again.

It can be, you have to try it several times.
Posted By: Shideravan

Re: SVN builds - new driver flood - 10/04/09 02:20 PM

Anna, yoou're using a virtual HD?
Did you created it with IMGtool?
To make a floppy img/dsk what program trhat you used?
IMGtool again?
Thank you!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/09 04:22 PM

Originally Posted By Shideravan
Anna, yoou're using a virtual HD?
Did you created it with IMGtool?
To make a floppy img/dsk what program trhat you used?
IMGtool again?
Thank you!


The harddisk emulation for any PC or compatible driver is broken, since the dma devices was fixed in version 0.133
Judge maybe will fix it.
I use a external tool called WinImage to create or edit PC disk images.
For the chd images, I use chdman.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/04/09 05:20 PM

It seems my SVN is locked.
I delete SVN\Bin\mess\ complete.
I tried :
Code:
svn unlock svn://messdev.fdns.net/mess/

and than
Code:
svn checkout svn://messdev.fdns.net/mess/

but still no running, have still a lock. frown

Posted By: judge

Re: SVN builds - new driver flood - 10/04/09 06:18 PM

Try doing an 'svn cleanup'
Posted By: Stiletto

Re: SVN builds - new driver flood - 10/04/09 08:41 PM

Originally Posted By Anna Wu
xtvga driver (still in progress)
Game : Prince of Persia (Floppy Disk)



Wow, sweet!

- Stiletto
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/05/09 02:13 AM

SVN r5967
CD-i driver (in progress)



Video: (note: is all drawn with a 4/8bpp BMP plane, no MPEG support has been added, nor has DYUV) http://www.youtube.com/watch?v=qC5AosaHbAU
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/05/09 02:22 PM

Originally Posted By judge
Try doing an 'svn cleanup'


Thank you, judge. smile
Posted By: Stiletto

Re: SVN builds - new driver flood - 10/05/09 02:23 PM

That's awesome. Congratulations JD (and CD-i Fan)!

- Stiletto
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/05/09 02:38 PM

SVN r5974
MikroMikko 1 M6
Nokia CP/M 2.2



Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/06/09 02:29 AM

SVN r5967
cdimono1 driver (still in progress)
Game: Frog Feast (Demo)

Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/06/09 02:56 AM

And the first non-Philips-logo screen of Striker Pro:



There seem to be a lot of graphical issues in the intro, though, so this should give me a good test case. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/06/09 02:59 AM

Great JD ! smile
Many thanks goes to CD-i Fan too.
Posted By: Stiletto

Re: SVN builds - new driver flood - 10/06/09 03:18 AM

Originally Posted By Just Desserts
And the first non-Philips-logo screen of Striker Pro:



There seem to be a lot of graphical issues in the intro, though, so this should give me a good test case. smile


Christ, you're fast! Kudos...

- Stiletto
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/06/09 03:22 AM

Originally Posted By Stiletto
Originally Posted By Just Desserts
And the first non-Philips-logo screen of Striker Pro:



There seem to be a lot of graphical issues in the intro, though, so this should give me a good test case. smile


Christ, you're fast! Kudos...

- Stiletto


Fast nothing, it already worked that far, I just hadn't tested it yet. smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/06/09 03:54 AM

SVN r5980
cdimono1 driver (still in progress)
Game: Striker Pro

http://www.youtube.com/watch?v=NeA2OCb-CaQ

Aaaaand I'm spent. Time for bed.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/06/09 04:04 AM

Originally Posted By Just Desserts
SVN r5980
cdimono1 driver (still in progress)
Game: Striker Pro

http://www.youtube.com/watch?v=NeA2OCb-CaQ

Aaaaand I'm spent. Time for bed.


Wow !

Take a rest now, JD.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/07/09 03:54 AM

SVN r5990
gbcolor driver
Now games like The Legend of Zelda - Link's Awakening DX are working.
Thank you for the fix, judge smile


Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/07/09 06:30 PM

SVN r5997
Amstrad PPC512/PPC640 driver (still in progess, video output and keyboard input need a fix)
Game : Marble Madness (I compress the .exe to have enough space on the 720K disk.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/08/09 01:09 PM

SVN r6013
[FMTMARTY] Re-organised ROM loading so that it runs valid boot code.



Hello Barry,

if I run the Xe emulator which use the same 2 roms but as one merged rom (SH1:5aaef45294eeeadeabdd5ecd92119fd5afa827ab/CRC32:39d81a66), it show me additional a boot intro.



I found a video
Maybe different roms and boot screens between Marty and Marty 2 ?

The Xe emulator also support the CarMarty rom (merged)(SH1:9c210d41eb430bc37248ae0f10b3d8c10ffe0080/CRC32:8ce1de19)

Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/08/09 03:53 PM

Do we know what type the CD-ROM is on the FM-Towns and/or Marty? If it's looking for 0xeb and 0x14 in two of the ports on startup that's ATAPI, for instance.
Posted By: incog

Re: SVN builds - new driver flood - 10/08/09 08:02 PM

Oh awesome, it's great to see that the Car Navi Marty is already dumped.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/08/09 09:00 PM

I believe that it's some form of SCSI-1 controller, but don't have any more info on it other than what's in the FM-Bochs source.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/09/09 06:52 AM

It is very difficult to find more details about the CD-ROM drive and the controller.
Guru dumped already the roms from the Marty (not CarMarty) console. Maybe he still have the console and can give us more details ?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/09/09 08:21 PM

Hmm. The FM-Bochs source doesn't look like SCSI-1, it's probably custom.

OT: two wiki things. There's 2 of every system on the "drivers" page, and it'd be cool if each parent driver page included all the screenshots available for that system.
Posted By: Duke

Re: SVN builds - new driver flood - 10/09/09 10:06 PM

Yes, I'm working on completely new wiki system pages. This will be the base page: http://mess.redump.net/systems
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/09/09 10:10 PM

Very nice!
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/10/09 05:18 AM

Originally Posted By R. Belmont
Hmm. The FM-Bochs source doesn't look like SCSI-1, it's probably custom.


Yes, looks that way. I've found a list of IRQs for the FM-Towns, and it shows the SCSI controller and CD-ROM controller to have different IRQs (SCSI is IRQ8, CD-ROM is IRQ9). SCSI is likely to be used for HDD, something the system doesn't require (most software boots from either floppy or CD-ROM). There's also Linux/Towns source for me to look through to get a decent idea some more of the functionality.
Posted By: Guru

Re: SVN builds - new driver flood - 10/10/09 05:30 AM

Originally Posted By Anna Wu
It is very difficult to find more details about the CD-ROM drive and the controller.
Guru dumped already the roms from the Marty (not CarMarty) console. Maybe he still have the console and can give us more details ?


no, I was sent only the PCB. I dumped the surface mount ROMs on it and did a readme for the PCB then returned it to the XE Emulator guy.
Posted By: Guru

Re: SVN builds - new driver flood - 10/10/09 05:47 AM

Originally Posted By incog
Oh awesome, it's great to see that the Car Navi Marty is already dumped.


but not public which means we'll have to re-dump it. None of the XE Emulator ROMs are public, sadly, except the ones I dumped.

Posted By: etabeta78

Re: SVN builds - new driver flood - 10/10/09 07:55 AM

Originally Posted By Duke
Yes, I'm working on completely new wiki system pages. This will be the base page: http://mess.redump.net/systems


cool!!!! that's exactly the kind of page I dreamed of when I started bugging you about it!!!! wink

I hope you managed to solve all the wiki coding quirks that were giving you headaches
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/10/09 12:12 PM

Err, doesn't Anna have the Car Marty ROMs or was the screenshot she posted from somewhere else?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/10/09 12:35 PM

The Car Marty ROMs was dumped by using the tool " Rombeger " from the UNZ author :
Quote:
rem USAGE: rombeger [ROM No.0-9] [drive]
rem ROM No.0: fmt_sys.rom(256KB)
rem ROM No.1: fmt_fnt.rom(256KB)
rem ROM No.2: fmt_dos.rom(512KB)
rem ROM No.3: fmt_dic.rom(512KB)
rem ROM No.4: fmt_f20.rom(512KB)
rem ROM No.5: mar_ex0.rom(512KB)
rem ROM No.6: mar_ex1.rom(512KB)
rem ROM No.7: mar_ex2.rom(512KB)
rem ROM No.8: mar_ex3.rom(512KB)
rem ROM No.9: mytowns.rom(32BYTE)
rem [drive] 0:current 1:a:\ 2:b:\
rombeger 0 1
rombeger 1 1
rombeger 2 1
rombeger 3 2
rombeger 4 2
rombeger 9 2


I made the screenshot from the Xe emulator.
Not sure it is a good dump or not but it works on the Xe emulator.
You can check the the ROMs for the FM-Towns, I think they was dumped with Rombeger too.
Barry will check the roms. wink
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/10/09 12:54 PM

Ahh, good smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/10/09 01:11 PM

SVN r6028
NES driver
Game : Super Pang (Cartridge, .unf)

Thanks etabeta78, for your improvements. smile



Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/10/09 01:15 PM

Originally Posted By Duke
Yes, I'm working on completely new wiki system pages. This will be the base page: http://mess.redump.net/systems


This is very helpful ! smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/10/09 10:42 PM

WIP: Macintosh IIx (68030) with RasterOps 264 video card showing the video card's BIOS screen. If I'm not mistaken, this is the first time ever that a Mac emulator has shown video from a properly emulated NuBus video card instead of just shunting a fake driver into native code ;-)

Posted By: Duke

Re: SVN builds - new driver flood - 10/11/09 12:46 AM

Wow, thats awesome smile
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 10/11/09 12:52 AM

I believe you really are the first, too. It's a long time since I've seen that screen - it brings back memories.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/11/09 02:10 AM

Here's a screenshot from real hardware (actually the SE/30 PDS variant of the board, but it looks like it works quite similarly).

The SE/30 motherboard I bought arrived today and I successfully dumped the video declaration ROM from it (it's separate from the main system ROM SIMM and contains mostly credits for the SE/30 hardware since the driver is very tiny). The emulated SE/30 detects the display and puts up the familiar Mac "rug pattern" but doesn't go farther yet.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/11/09 04:24 AM

Alright. Supporting the next video mode the RasterOps board switches into nets us this:



VRAM layout in 1 bit per pixel mode on that card is nearly Apple II hi-res levels of weird. Also, it doesn't like my harddisk images created on the (emulated) Mac Plus for reasons I need to investigate (I see where in the code it's rejecting them, I just don't know the details yet), and nobody has a clue how the SWIM works so no floppies pending a major deconstruction of the .Sony driver. Whee smile

ETA: declaration ROM dumps and chip lists for other Mac II NuBus video cards accepted, PM me ;-)
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 10/11/09 07:35 AM

Hang on, isn't the SWIM just like the Apple II disk drive circuitry (the "Woz Machine"), only integrated into a single chip?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/09 08:10 AM

SVN r6034

Is the Famicom Disk support broken ?
If i try to load a fds file I get a crash.
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/11/09 08:46 AM

@Anna Wu: whooops... I forgot 1 line in the NES commit yesterday

I'll fix famicom soon. thanks for testing


@Arbee: wonderful news. I've never seen those macintosh machines at time and I'm quite curious to see them in action!

btw, could your HD problem be related to the following crash I was getting on SDLMESS? http://bugzilla.mess.org/show_bug.cgi?id=1664
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/09 10:01 AM

Found a short info about the SCSI controller for the FMV-Towns. Unfortunately no more details.
Fujitsu is also talking about this controller.

Maybe the FMV-Towns is a newer model as FM-Towns.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/11/09 10:26 AM

The FM V-Towns is a higher spec version of the FM-Towns. Probably best to try to get the simpler systems done first, I think smile

Anyway, I think I've done all I can for the Towns driver for now, since i386 protected mode bugs are making more of an impact now. It's impossible to call interrupts in protected mode since the vector addresses are 8 bytes long, and the MAME i386 core treats them as 4 bytes each, thusly calling the wrong vector. It does at least use the IDT, I suppose that's something. smile Also, the I/O priviledge level and nested task flags can't be set. Damn shame that the Towns version of MS-DOS requires protected mode, I was hoping that it didn't.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/09 10:32 AM

Maybe you will find a way. smile
Can not wait to load a disk successful and later a cd image.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/11/09 10:33 AM

Vas: IWM ("Integrated Woz Machine") is the single-chip Woz controller. It can handle 400k and 800k disks (plus Lisa twiggy drives and Apple II/III 5.25"), all of which are GCR.

SWIM ("Super Woz Integrated Machine") adds MFM capability and can do 720k MFM, 800k GCR, or 1.44M MFM (when connected to what Apple called "SuperDrive" at the time). The registers are largely identical to the IWM (NetBSD is capable of driving the SWIM/SuperDrive in 800k back-compatibility mode only) but it's sufficiently different that just slapping an IWM+800k drive there makes the firmware quite unhappy on SWIM-equipped Macs.

The original Mac II did have an IWM + 800k drive and that might be our way in, but that particular ROM is very hard to find. (Later-model IIs had the SWIM+SuperDrive just like later-model SEs did, and the upgrade uses the same firmware that the IIx, IIcx, and SE/30 did).


eta: that's unrelated. Everything I have shows block 0 as complete and valid (the recently released HFSExplorer, not to be confused with the old HFVExplorer, opens -extract'd CHDs fine). Just the newer firmware doesn't like it for some reason.
Posted By: Justin

Re: SVN builds - new driver flood - 10/11/09 10:39 AM

src\mess\machine\nes_mmc.c(7509) : warning C4717: 'mapper153_w' : recursive on all control paths, function will cause runtime stack overflow
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/09 10:44 AM

Originally Posted By Anna Wu
SVN r6034

Is the Famicom Disk support broken ?
If i try to load a fds file I get a crash.


SVN r6037
[FAMICOM] Fixed Disk System
I can now load a fds file without crash but I get a disk trouble err. 23 in emulation.


Posted By: etabeta78

Re: SVN builds - new driver flood - 10/11/09 10:50 AM

@Arbee: thanks for the answer

@Justin & Anna: tonight I'll try to fix both problems
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/09 10:51 AM

Originally Posted By etabeta78
@Arbee: thanks for the answer

@Justin & Anna: tonight I'll try to fix both problems


Thank you to take care about the problems. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/11/09 12:46 PM

both bugs are now fixed locally. I will commit the changes to svn tonight, hopefully with some other progress smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/11/09 03:27 PM

Mac SE/30 internal video now working, although between this and the previous picture I'm clearly missing something about how NuBus maps stuff smile


Posted By: Stiletto

Re: SVN builds - new driver flood - 10/11/09 04:07 PM

Originally Posted By Anna Wu
It is very difficult to find more details about the CD-ROM drive and the controller.
Guru dumped already the roms from the Marty (not CarMarty) console. Maybe he still have the console and can give us more details ?


I was at a videogame convention yesterday - one of the sellers had an FM Towns Marty for sale - $800, so I decided to pass on it. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/09 04:22 PM

Here are some pictures about a disassembled Marty. Unfortunately the pictures are not big enough to take a closer look, especially for the chips .
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/11/09 04:44 PM

Fixed the byte order for the Mac SE/30 on-board video smile



...and a similar fix sorts out the RasterOps board on the IIx.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/11/09 04:48 PM

Great !

The SVN is waiting ... smile
Posted By: Stiletto

Re: SVN builds - new driver flood - 10/11/09 06:42 PM

Originally Posted By Anna Wu
Here are some pictures about a disassembled Marty. Unfortunately the pictures are not big enough to take a closer look, especially for the chips .


Indeed, so what we can scavenge from the pictures and the posts there:
http://nfggames.com/forum2/index.php?topic=2130.0
and there: http://assemblergames.com/forums/archive/index.php/t-12771.htmlp/t-14904.html
and another place I cannot link to...

Is:
FM Towns Marty
CPU: AMD DS39473A
OSC2: 32 MHz (1/2 divider makes the CPU 16 MHz)
OSC1: 28.63636
Video: CE31755, custom, 160-pin
Video DAC: MB40968V
Sound: Yamaha YM3xxx (likely YM3438 OPN2C - and Barry already has Ricoh RF5c68 in the driver)
VRAM: 512kB in four 128kB TMS48C121DZ chips
Sprite RAM: 128kB Sprite RAM in one TC511664BJ-80 chip
ROM: 2x Hitachi HN624116 (256KB each) and 6x or 7x 512KB unknown
and FM Towns Marty 2 has those same specs.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/12/09 04:18 AM

r6045: CD-i Mono I driver: Zenith

http://www.youtube.com/watch?v=4CeN2SBti-8
Posted By: Kaylee

Re: SVN builds - new driver flood - 10/12/09 10:43 AM

I'm trying to compile r6049. and it crashes when trying to compile/generate M86K Source Files.

Throws out the following:


M86kcpu.c: in function 'readword_d32_mmu':
m68kmmu.h:15: warning: 'tamode' may be used uninitialized in this function
m68kmmu.h:15: note: 'tamode' was declared here

does the same with: 'tbl_entry' and 'tbmode'

then goes to:
mingw32-make: ***[obj/windows/mess/emu/cpu/m68000/m68kcpu.o] Error 1

I always del src/mess/obj before each compile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/12/09 12:54 PM

That's actually a bug in GCC (it's not recognizing that fatalerror() halts execution) that was fixed in 4.4.1 (which is why I don't see it on Fedora 11) but SVN #6050 should work around it.
Posted By: Kaylee

Re: SVN builds - new driver flood - 10/12/09 02:19 PM

Thanx the Update works
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/12/09 08:53 PM

Originally Posted By Anna Wu
Originally Posted By Anna Wu
SVN r6034

Is the Famicom Disk support broken ?
If i try to load a fds file I get a crash.


SVN r6037
[FAMICOM] Fixed Disk System
I can now load a fds file without crash but I get a disk trouble err. 23 in emulation.




Thanks for the final fix, etabeta78 smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/14/09 07:16 AM

SVN r6067
[CD-I] Fixed backdrop/cursor colors for consistency, mouse simulation, and CDIC behavior.

This fixed up the Lemmings intro a bit: http://www.youtube.com/watch?v=ctdN5MSJ2sE
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/09 08:45 AM

I have a question. I want write a avi file.
If I use only the avi file name, it will saved in in snap folder.
If I use the avi file name with path, nothing will be saved on the choosed path.

Example:

mess nes -r 640x480 -window -nodebug -nolog -nosound -aviwrite Demo.avi -cart "c:\Emulator\NES\Games\Zelda2.zip"

Will be saved: c:\Emulator\Mess\Mess SVN\Snap\Demo.avi

mess nes -r 640x480 -window -nodebug -nolog -nosound -aviwrite c:\Demo.avi -cart "c:\Emulator\NES\Games\Zelda2.zip"

Nothing will be saved on c:\

What is wrong ?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/14/09 08:47 AM

Originally Posted By Just Desserts
SVN r6067
[CD-I] Fixed backdrop/cursor colors for consistency, mouse simulation, and CDIC behavior.

This fixed up the Lemmings intro a bit: http://www.youtube.com/watch?v=ctdN5MSJ2sE


SVN r6071
[CDI] Added half-done CLUT7+7 support, enough for Tetris to become fully playable.
Game: Tetris

http://www.youtube.com/watch?v=QyAkiOcxzIA
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/14/09 08:47 AM

Originally Posted By Anna Wu
I have a question. I want write a avi file.
If I use only the avi file name, it will saved in in snap folder.
If I use the avi file name with path, nothing will be saved on the choosed path.

Example:

mess nes -r 640x480 -window -nodebug -nolog -nosound -aviwrite c:\Demo.avi -cart "c:\Emulator\NES\Games\Zelda2.zip"

What is wrong ?


that you cannot save avi outside the snap folder. you can only choose the filename
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/09 08:50 AM

Thank you for your answer. So it was not my mistake. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/09 08:56 AM

Originally Posted By Just Desserts
Originally Posted By Just Desserts
SVN r6067
[CD-I] Fixed backdrop/cursor colors for consistency, mouse simulation, and CDIC behavior.

This fixed up the Lemmings intro a bit: http://www.youtube.com/watch?v=ctdN5MSJ2sE


SVN r6071
[CDI] Added half-done CLUT7+7 support, enough for Tetris to become fully playable.
Game: Tetris

http://www.youtube.com/watch?v=QyAkiOcxzIA


Good work, JD smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/14/09 10:42 AM

SVN r6066
x68kxvi driver (in progress)
Game : Marble Madness and Scorpius (Floppy Disk)



x68030 driver (in progress)


Posted By: Robbbert

Re: SVN builds - new driver flood - 10/14/09 10:59 AM

Anna: sounds similar to this mametesters bug:

http://www.mametesters.org/view.php?id=3437
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/14/09 11:00 AM

which is what I said (you cannot decide the full path, but only the avi name) and which could be not a bug but by design wink
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/15/09 02:40 PM

Yummy, tasty videos:

Lemmings, full intro: http://www.youtube.com/watch?v=UvJxhwzivuc
Hotel Mario: http://www.youtube.com/watch?v=CzIMAacYFi8
Whack-A-Bubble: http://www.youtube.com/watch?v=nAxiBlcLvBM
Zenith: http://www.youtube.com/watch?v=CL2ELKCGl88
Pac-Panic: http://www.youtube.com/watch?v=Y-S99nHIgh0
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/15/09 04:24 PM



Fantastic smile

I hope, the audio/sound support will coming soon ...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/15/09 05:02 PM

Just tried Super Mario´s Wacky Worlds on the cdimono1 driver.


Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/17/09 10:04 AM

Merlin's Apprentice: http://www.youtube.com/watch?v=WXM8GhLEnd0

I think the music quality is all the more reason for us to fix up the audio emulation - without the glitches, I'll bet the music sounds really damn good. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/17/09 10:21 AM

Originally Posted By Just Desserts
Merlin's Apprentice: http://www.youtube.com/watch?v=WXM8GhLEnd0

I think the music quality is all the more reason for us to fix up the audio emulation - without the glitches, I'll bet the music sounds really damn good. smile


Amazing, the emulation goes better and better. laugh
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/17/09 08:57 PM

SVN r6116: CD-i Mono-I driver
Fixed Level B, C, and mono Level A sound.

Merlin's Apprentice: http://www.youtube.com/watch?v=8K4vxFk2cw0
Link: Faces of Evil: http://www.youtube.com/watch?v=AlPZ3moteRs MY BOOOOOOOOY
Zelda: The Wand of Gamelon: http://www.youtube.com/watch?v=Gk2rjtG4ZYg
Hotel Mario: http://www.youtube.com/watch?v=DYl24XVyJKI
Lemmings: http://www.youtube.com/watch?v=7tQqpBqT50A
Tetris: http://www.youtube.com/watch?v=0hT25Pz2LnU
Zenith: http://www.youtube.com/watch?v=_5ACN6RGz6I
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/18/09 06:05 AM

With SVN r6126, you might notice:

- CD-i Backgammon is playable
- Caesar's Palace is playable
- Connect 4 is playable
- CD-i Pinball is playable
- The Wacky World of Mini Golf is nearly playable (obstacle timing issues)
- Other Capitol Disc Interactive games (that Tangram game, for instance) run farther
- The Zelda games will show the full title animation rather than hanging

Enjoy!
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 11:58 AM

SVN r6136
Memotech MTX 500/512 driver

I have still problems to load successful snapshots (.mtb)
The sysinfo.dat is talking about the tape usuage which not exist for these driver.
Posted By: Robbbert

Re: SVN builds - new driver flood - 10/19/09 02:13 PM

Posted By: etabeta78

Re: SVN builds - new driver flood - 10/19/09 02:16 PM

that is COOL!!!
Posted By: Stiletto

Re: SVN builds - new driver flood - 10/19/09 02:39 PM

Congrats, boys...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 04:29 PM

@robbbert

Can you play the game on the last SVN release ?
Because if I start the same game with the last official MESS version I get the same title screen. Unfortunately, the game is still not playable.

Exist any other game which can show the title screen without graphical errors on the last SVN release ?
Posted By: judge

Re: SVN builds - new driver flood - 10/19/09 04:44 PM

The games are still not playable.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 04:46 PM

Originally Posted By judge
The games are still not playable.


Thanks for the answer. smile
Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 04:47 PM

Hello all,
When i want to select a driver in gui with last svn6138, the emulator crash, you know why i have this error please?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 04:52 PM

Maybe you need a clean compiling ? I had the same issue in the past (SVN r6098).
Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 04:54 PM

What mean by "clean compliling"? Erase obj folder for to make a new compil?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/19/09 04:58 PM

"make clean" for SDL, "mingw32-make clean" for Windows then rebuild as usual.
Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 05:00 PM

ok i test it now, thanks for help all smile
Just another little question, how to make for to boot game with cdi driver? Iso can work or need to be converted to chd format?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 05:09 PM

Originally Posted By Multipass
ok i test it now, thanks for help all smile
Just another little question, how to make for to boot game with cdi driver? Iso can work or need to be converted to chd format?


On MESS you can only use cd images in chd format.
Here a example (1-3), how i create a chd image.
Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 05:12 PM

Somes options will be checked in burn at once or default options only?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 05:14 PM

Originally Posted By Multipass
Somes options will be checked in burn at once or default options only?


Read Settings :
Read RAW (marked)



Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 05:17 PM

I have "Raw RW subs", and "Simple toc" checked, its ok too?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 05:20 PM

Originally Posted By Multipass
I have "Raw RW subs", and "Simple toc" checked, its ok too?


I no need.

No Checking
No Subs
Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 05:24 PM

I have tested clean compil and create chd, all work now! More more thanks for your help all! smile wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/19/09 05:27 PM

Originally Posted By Multipass
I have tested clean compil and create chd, all work now! More more thanks for your help all! smile wink


Wow, you must have a fast PC.
Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 05:53 PM

Yes Yes i have a good Pc in my house smile 3ghz, 2gb ram , intel quad core duo
Posted By: Sune

Re: SVN builds - new driver flood - 10/19/09 06:28 PM

Me wants a Quad Core Duo too.

Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 06:41 PM

Someone know if Philips CDI have an equivalent in arcade please?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/19/09 06:46 PM

Nope. The core chips were used in a few obscure gambling games but none of them were the full CD-i system with OS9 and everything.
Posted By: Multipass

Re: SVN builds - new driver flood - 10/19/09 06:48 PM

Ok more thanks R.Belmont wink
Posted By: Kale

Re: SVN builds - new driver flood - 10/19/09 07:10 PM

http://cdii.blogspot.com/2009/02/arcade-machines-based-on-cd-i.html

Actually yes, there's an arcade game named Quizard that's actually a more-or-less standard Philips CD-i with coin inputs.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/19/09 07:17 PM

Ok, that *is* a standard CD-i consumer unit smile There was never any real CD-i based JAMMA board or anything though.
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 10/20/09 06:56 AM

Originally Posted By R. Belmont
There was never any real CD-i based JAMMA board or anything though.

There exists something called the "JNMS player", supposedly a very early prototype (even older then the CDI 18x series), but it is probably not related.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/09 09:50 AM

SVN r6141
cdimono1 driver (in progress)
Game : Super Mario`s Wacky Worlds

Great, now I have sound in game. The sound is stuttering a bit but ok. I am sure, a improvement will come. smile
Posted By: Curt Coder

Re: SVN builds - new driver flood - 10/20/09 03:51 PM

Bondwell 12/14 floppy loading is broken once again frown
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/20/09 03:59 PM

I could be wrong, but I fear it had been broken since long before the floppy changes by Micko, back until some older NEC FDC changes

EDIT: nope. I was thinking to this problem
http://www.bannister.org/forums/ubbthreads.php?ubb=showflat&Number=53603

and it had been fixed. hence the regression is more recent
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/09 04:57 PM

Originally Posted By etabeta78
I could be wrong, but I fear it had been broken since long before the floppy changes by Micko, back until some older NEC FDC changes

EDIT: nope. I was thinking to this problem
http://www.bannister.org/forums/ubbthreads.php?ubb=showflat&Number=53603

and it had been fixed. hence the regression is more recent


Micko fixed the boot problem for the bw2 driver which is still working.
The bw12/bw14 driver have now a boot problem again.
It was working before.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/09 06:48 PM

SVN r6149
cdimono1 driver (in progress)

I think the sound was better with SVN r6143 in some games like Space Ranger.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/20/09 07:01 PM

Originally Posted By Anna Wu
SVN r6149
cdimono1 driver (in progress)

I think the sound was better with SVN r6143 in some games like Space Ranger.


Could you be more specific? I have listened to the audio in Space Ranger between r6143 and r6149, and did not notice any difference.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/09 07:13 PM

Originally Posted By Just Desserts
Originally Posted By Anna Wu
SVN r6149
cdimono1 driver (in progress)

I think the sound was better with SVN r6143 in some games like Space Ranger.


Could you be more specific? I have listened to the audio in Space Ranger between r6143 and r6149, and did not notice any difference.


Unfortunately I have no more the r6143 to compare in detail. frown
Maybe my feeling is subjective to think the sound was better in animation (small screen) before you start the mission. Also the music in gameplay.
Posted By: judge

Re: SVN builds - new driver flood - 10/20/09 07:31 PM

to update your source tree to revision 6143 do:
Code:
svn update -r6143
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/09 07:39 PM

Originally Posted By judge
to update your source tree to revision 6143 do:
Code:
svn update -r6143


Thank you, i tried without success.

Posted By: etabeta78

Re: SVN builds - new driver flood - 10/20/09 07:41 PM

I think you must be in you mess folder and launch svn from there (either adding H:\svn\bin to PATH, or launching H:\svn\bin\svn.exe update -r6143 )

but it's been quite some time since I last used svn in windows (without tortoisesvn), hence I could be wrong
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/09 07:46 PM

Originally Posted By etabeta78
I think you must be in you mess folder and launch svn from there (either adding H:\svn\bin to PATH, or launching H:\svn\bin\svn.exe update -r6143 )

but it's been quite some time since I last used svn in windows (without tortoisesvn), hence I could be wrong


This info was very helpful and it works. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/20/09 08:26 PM

Originally Posted By Just Desserts
Originally Posted By Anna Wu
SVN r6149
cdimono1 driver (in progress)

I think the sound was better with SVN r6143 in some games like Space Ranger.


Could you be more specific? I have listened to the audio in Space Ranger between r6143 and r6149, and did not notice any difference.


I think you are right. Here the comparison between SVN r6143 and SVN r6149 and the Space Ranger sound.

mess cdimono1 -r 640x480 -window -nonewui -nodebug -nolog -wavwrite sr.wav -bios 0 -cdrom "C:\Emulator\Mess\Software\CD-I\CDROM\Space Ranger.chd"
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/21/09 02:22 AM

Originally Posted By Curt Coder
Bondwell 12/14 floppy loading is broken once again frown

Fixed by removing nec765_idle() from the seek setup function. It was making it think that the recalibrate had finished when it hadn't, thus making the FDC treat Sense Drive Status as an invalid command. I tested the PCW driver after the change and it seems to still be working, post a reply if it breaks something else.
Posted By: Micko

Re: SVN builds - new driver flood - 10/21/09 06:57 AM

Only thing noticed not to work now is Elwro 800, that line was in Mariusz patch for Elwro. So here something is definitly not ok.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/21/09 07:23 AM

Well, the Bondwell checks the main status register, and if either bit 4 or 5 is set, then it waits and checks again, until neither bit is set. Bit 5 is execution mode, and should be set high during a seek/recalibrate operation. nec765_idle() clears both bits.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/21/09 08:17 AM

Ok, appears as though the FDC Busy flag is set, which it as I understand it, shouldn't be during a seek or recalibrate. Having the busy flag in the main status register reset at the beginning of a seek/recalibrate gets elwro800 running, and booting CP/J. Bondwell 12/14 still works, and I tested the PCW, CPC and X68000 drivers which also still appear to work. Reply if you know of any drivers broken by this change.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/21/09 08:33 AM

SVN r6152

BW12/BW15 boot disks : working
(Still can not use the keyboard input but this is another problem)

Elwro boot disks (CP/J, Amstrad) boot disks : working

PCW8256 boot disks : working
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/21/09 09:47 AM

Originally Posted By Anna Wu
SVN r6152

BW12/BW15 boot disks : working
(Still can not use the keyboard input but this is another problem)

Elwro boot disks (CP/J, Amstrad) boot disks : working

PCW8256 boot disks : working


EDIT: Found the reason, I was not waiting long enough. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/21/09 05:37 PM

r6163/r6164
It seems these revisions produce a compiling error.
Posted By: judge

Re: SVN builds - new driver flood - 10/21/09 05:57 PM

Fixed
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/21/09 06:13 PM

Originally Posted By judge
Fixed


Thank you. smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/23/09 11:22 PM

Some MESS N64 WIP. Since it looks like the N64 driver is going to stagnate in legal hell without external intervention, I took it upon myself to port over a bunch of fixes from Mupen64. Things are much better, but there are still a lot of Z-sorting, blending, S/T lookup, audio and CPU bugs. frown

Super Mario 64: http://www.youtube.com/watch?v=LD729BKpKEE
Tetrisphere: http://www.youtube.com/watch?v=KzVk7z1oXIM
Mario Kart 64 (bad audio): http://www.youtube.com/watch?v=DgO6dSe4_jk
The New Tetris: http://www.youtube.com/watch?v=9YhfPSfsyKM
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/24/09 07:52 AM

very cool. I love the fact that the big screen at the entrance of Luigi Raceway tunnel in MK64 works with no apparent issue! thumbs up!

but... where are Mario's mustaches?!? wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/24/09 08:43 AM

Originally Posted By Just Desserts
Some MESS N64 WIP. Since it looks like the N64 driver is going to stagnate in legal hell without external intervention, I took it upon myself to port over a bunch of fixes from Mupen64. Things are much better, but there are still a lot of Z-sorting, blending, S/T lookup, audio and CPU bugs. frown

Super Mario 64: http://www.youtube.com/watch?v=LD729BKpKEE
Tetrisphere: http://www.youtube.com/watch?v=KzVk7z1oXIM
Mario Kart 64 (bad audio): http://www.youtube.com/watch?v=DgO6dSe4_jk
The New Tetris: http://www.youtube.com/watch?v=9YhfPSfsyKM


I wish to have a faster PC. frown Your N64 sound is much better.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/24/09 08:59 AM

Originally Posted By Anna Wu
I wish to have a faster PC. frown Your N64 sound is much better.


-aviwrite writes every frame, so even extremely slow emulation is full speed when played back... smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/24/09 09:03 AM

Originally Posted By Just Desserts
Originally Posted By Anna Wu
I wish to have a faster PC. frown Your N64 sound is much better.


-aviwrite writes every frame, so even extremely slow emulation is full speed when played back... smile


Ahh, ok smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/24/09 11:41 AM

SVN r6178/9: CD-i driver: Ported over some fixes provided by CD-i Fan, Hotel Mario now goes in-game and works great. Other games no longer hang. Unfortunately, Lemmings still does not have music in-game, but it does play a few sectors' worth of music data before the music dies.

Hotel Mario: http://www.youtube.com/watch?v=-FbwA-CpsMo
Zombie Dinos: http://www.youtube.com/watch?v=Qg3bQrEmi3Y
Toon Jukebox: http://www.youtube.com/watch?v=ykGMynshsAY

This checkin pair also marks the upgrading of the driver's status from NOT_WORKING to IMPERFECT_GRAPHICS and IMPERFECT_SOUND. I will be more willing to listen to bug reports now, please just make sure that the disc rip is correct, and that it isn't a DVC-requiring game.

Known issues so far include:
Merlin's Apprentice: Perfect
Namco Arcade Classics: Hard to play with a joystick, but otherwise perfect
CD-i Backgammon: Perfect
Battleship: Minor graphical glitch at the start of a match, otherwise perfect
CD Shoot: Perfect
Classic Jukebox: Perfect
Compton's Interactive Encyclopedia 1992: Perfect
Caesar's Palace Casino: Perfect?
CD-i Pinball: Perfect?
Connect Four: Perfect?
Christmas Crisis: Requires DVC
Escape from Cyber City: Hangs after boot video
Create Your Own Charicature: Perfect?
Dark Castle: Perfect?
Link - Faces of Evil: Perfect?
Zelda - The Wand of Gamelon: Perfect?
Golden Oldies 1: Perfect
Golden Oldies 2: Perfect
Hotel Mario: Perfect?
International Tennis Open: Hangs after boot video
Jeopardy: Perfect?
Jigsaw - The Ultimate Electronic Puzzle: Graphically perfect, but mouse pointer never responds
Kether: Perfect?
Laser Lords: Seems to lack certain controls - enter/exit building, talk to people - but otherwise perfect
Lemmings: No background music, but otherwise perfect
Mario Wacky Worlds (Prototype): Perfect?
Mega Maze: Hangs when trying to go in-game
Micro Machines: Perfect?
The Wacky World of Mini-Golf: Perfect?
Myst: Perfect?
Name That Tune: Perfect?
Pac-Panic: Perfect (no sound?)
Sargon Chess: Perfect (a bit laggy?)
Richard Scarry's Best Neighbor Book Ever (or whatever it's called): Perfect?
Space Ranger: Perfect?
Striker Pro: Perfect?
Tangram - The Ultimate Chinese Game: Perfect?
Tetris: Perfect
Text Tiles: Perfect?
Toon Jukebox: Perfect
Whack-A-Bubble: Perfect?
Zelda's Adventure: Hangs during the intro
Zenith: Perfect?
Zombie Dinos: Perfect?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/24/09 01:15 PM

JD, when you make a rest ? I have the feeling you are working 24 hours. smile

PS: I will test the last changes.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/24/09 02:17 PM

SVN r6178/6179
CD-i driver

Wow, Dark Castle is now working without a crash. smile
Still have problems with cdiemu 0.52, if i choose a door/level in the great hall.

Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/24/09 04:05 PM

In other news, here's a fun little montage, part one of two: http://www.youtube.com/watch?v=nUXdmWfhZl4
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/24/09 04:25 PM

Originally Posted By Just Desserts
In other news, here's a fun little montage, part one of two: http://www.youtube.com/watch?v=nUXdmWfhZl4


Nice smile
Posted By: Sune

Re: SVN builds - new driver flood - 10/24/09 07:04 PM

Haha, über corny soundtrack!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/25/09 01:03 AM

Getting a bit further with the newer Macs...

Posted By: ranger_lennier

Re: SVN builds - new driver flood - 10/25/09 05:53 AM

That's a lot of perfects! Nice work.

I have to admit I've never heard great things about CD-i games, but are there any I ought to check out? I have thought Mario's Wacky Worlds sounded interesting ever since I heard that it existed.

Originally Posted By Just Desserts

This checkin pair also marks the upgrading of the driver's status from NOT_WORKING to IMPERFECT_GRAPHICS and IMPERFECT_SOUND. I will be more willing to listen to bug reports now, please just make sure that the disc rip is correct, and that it isn't a DVC-requiring game.

Known issues so far include:
Merlin's Apprentice: Perfect
Namco Arcade Classics: Hard to play with a joystick, but otherwise perfect
CD-i Backgammon: Perfect
Battleship: Minor graphical glitch at the start of a match, otherwise perfect
CD Shoot: Perfect
Classic Jukebox: Perfect
Compton's Interactive Encyclopedia 1992: Perfect
Caesar's Palace Casino: Perfect?
CD-i Pinball: Perfect?
Connect Four: Perfect?
Christmas Crisis: Requires DVC
Escape from Cyber City: Hangs after boot video
Create Your Own Charicature: Perfect?
Dark Castle: Perfect?
Link - Faces of Evil: Perfect?
Zelda - The Wand of Gamelon: Perfect?
Golden Oldies 1: Perfect
Golden Oldies 2: Perfect
Hotel Mario: Perfect?
International Tennis Open: Hangs after boot video
Jeopardy: Perfect?
Jigsaw - The Ultimate Electronic Puzzle: Graphically perfect, but mouse pointer never responds
Kether: Perfect?
Laser Lords: Seems to lack certain controls - enter/exit building, talk to people - but otherwise perfect
Lemmings: No background music, but otherwise perfect
Mario Wacky Worlds (Prototype): Perfect?
Mega Maze: Hangs when trying to go in-game
Micro Machines: Perfect?
The Wacky World of Mini-Golf: Perfect?
Myst: Perfect?
Name That Tune: Perfect?
Pac-Panic: Perfect (no sound?)
Sargon Chess: Perfect (a bit laggy?)
Richard Scarry's Best Neighbor Book Ever (or whatever it's called): Perfect?
Space Ranger: Perfect?
Striker Pro: Perfect?
Tangram - The Ultimate Chinese Game: Perfect?
Tetris: Perfect
Text Tiles: Perfect?
Toon Jukebox: Perfect
Whack-A-Bubble: Perfect?
Zelda's Adventure: Hangs during the intro
Zenith: Perfect?
Zombie Dinos: Perfect?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/25/09 06:13 AM

Mario's Wacky Worlds is lame - I wouldn't even call it a prototype, more of an art test. You can load up worlds and walk left and right, and jump, but there's no interactivity with the environment at all, except sometimes you can stand on certain things. You can jump on enemies to kill them, but they'll just push you around if they run into you. There are no working coin boxes, no coins, no clearly-defined level exit conditions. Meh.

Honestly, Hotel Mario is addictive as all get out. It gets a bad rap because of the terrible cutscene animations, but the core gameplay elements are pretty fun overall. The background music is frantic and will get stuck in your head for hours after you play the game, so that's a plus as well.

Zenith is strange but addictive. I can't even liken it to any game that I've ever played. You have a marble, and you have to get to the goal. Your marble is infinitely bouncing "in" and "out" of the screen space, with your height being indicated by two shadows that are cast onto the ground, and the ground moving forward and back underneath your ball. You can move up, down, left and right to get around obstacles, or position your ball to pick up extra fuel, bonus balls, or points.

Whack-A-Bubble is a simplistic but fun game of Breakout, so that's good. It's also where I got the music for the first montage part - I'm actually having to resort to using non-CD-i music for part 2 because I can't find any other games with decently long enough music loops.

Merlin's Apprentice, I think, is the antidote to the animation found in the Zelda and Mario games. Although there were large areas of flat colors in all animations in order to be able to give an acceptable full-screen frame rate, Merlin's Apprentice proves that that is no excuse for bad animation, even moreso considering that most professional cartoons broadcast on TV during the same time frame used large areas of flat color as well.

Lemmings is a solid port, using the mouse to provide a more true-to-Amiga port than any other console port in existence, in my opinion.

Toon Jukebox is fun just to laugh at. Name That Tune the same, but for different reasons.

Originally Posted By ranger_lennier
That's a lot of perfects! Nice work.

I have to admit I've never heard great things about CD-i games, but are there any I ought to check out? I have thought Mario's Wacky Worlds sounded interesting ever since I heard that it existed.

Originally Posted By Just Desserts

This checkin pair also marks the upgrading of the driver's status from NOT_WORKING to IMPERFECT_GRAPHICS and IMPERFECT_SOUND. I will be more willing to listen to bug reports now, please just make sure that the disc rip is correct, and that it isn't a DVC-requiring game.

Known issues so far include:
Merlin's Apprentice: Perfect
Namco Arcade Classics: Hard to play with a joystick, but otherwise perfect
CD-i Backgammon: Perfect
Battleship: Minor graphical glitch at the start of a match, otherwise perfect
CD Shoot: Perfect
Classic Jukebox: Perfect
Compton's Interactive Encyclopedia 1992: Perfect
Caesar's Palace Casino: Perfect?
CD-i Pinball: Perfect?
Connect Four: Perfect?
Christmas Crisis: Requires DVC
Escape from Cyber City: Hangs after boot video
Create Your Own Charicature: Perfect?
Dark Castle: Perfect?
Link - Faces of Evil: Perfect?
Zelda - The Wand of Gamelon: Perfect?
Golden Oldies 1: Perfect
Golden Oldies 2: Perfect
Hotel Mario: Perfect?
International Tennis Open: Hangs after boot video
Jeopardy: Perfect?
Jigsaw - The Ultimate Electronic Puzzle: Graphically perfect, but mouse pointer never responds
Kether: Perfect?
Laser Lords: Seems to lack certain controls - enter/exit building, talk to people - but otherwise perfect
Lemmings: No background music, but otherwise perfect
Mario Wacky Worlds (Prototype): Perfect?
Mega Maze: Hangs when trying to go in-game
Micro Machines: Perfect?
The Wacky World of Mini-Golf: Perfect?
Myst: Perfect?
Name That Tune: Perfect?
Pac-Panic: Perfect (no sound?)
Sargon Chess: Perfect (a bit laggy?)
Richard Scarry's Best Neighbor Book Ever (or whatever it's called): Perfect?
Space Ranger: Perfect?
Striker Pro: Perfect?
Tangram - The Ultimate Chinese Game: Perfect?
Tetris: Perfect
Text Tiles: Perfect?
Toon Jukebox: Perfect
Whack-A-Bubble: Perfect?
Zelda's Adventure: Hangs during the intro
Zenith: Perfect?
Zombie Dinos: Perfect?
Posted By: Haze

Re: SVN builds - new driver flood - 10/25/09 09:26 AM

Burn Cycle works pretty well too, aside from some random graphic corruption, which appears to be self-correcting as it vanishes if you turn around again. The sound is also pretty messed up.

http://mamedev.emulab.it/haze/misc/burn_1.png
http://mamedev.emulab.it/haze/misc/burn_2.png
http://mamedev.emulab.it/haze/misc/burn_3.png
http://mamedev.emulab.it/haze/misc/burn_4.png

(yes, I left the LCD on just to see what it does)

Effacer is also playbable, although seems very laggy, it's ages before the buttons actually respond to anything.

http://mamedev.emulab.it/haze/misc/eff_1.png
http://mamedev.emulab.it/haze/misc/eff_2.png
http://mamedev.emulab.it/haze/misc/eff_3.png
http://mamedev.emulab.it/haze/misc/eff_4.png
http://mamedev.emulab.it/haze/misc/eff_5.png

to be honest, those DOS command line versions of hang-man are much more fun, wrapping up such a simple game with long drawn out sequences and speech only makes it worse.

Hieroglyph appears to want proper PAL modes

http://mamedev.emulab.it/haze/misc/hie_1.png
http://mamedev.emulab.it/haze/misc/hie_2.png

and again is a completely worthless game..

Inca seems to run ok too, there are some odd looking graphics on the left of the chase scene tho (shot 5) I also managed to kill the music by clicking too fast (maybe) in the puzzle.

http://mamedev.emulab.it/haze/misc/cdiinca_1.png
http://mamedev.emulab.it/haze/misc/cdiinca_2.png
http://mamedev.emulab.it/haze/misc/cdiinca_3.png
http://mamedev.emulab.it/haze/misc/cdiinca_4.png
http://mamedev.emulab.it/haze/misc/cdiinca_5.png
http://mamedev.emulab.it/haze/misc/cdiinca_6.png
http://mamedev.emulab.it/haze/misc/cdiinca_7.png
http://mamedev.emulab.it/haze/misc/cdiinca_8.png

also I couldn't figure out how to get past the first puzzle, it doesn't seem to match up with youtube ;/

Earth Command hangs during the intro video, or if you skip it.

http://mamedev.emulab.it/haze/misc/earth_1.png
http://mamedev.emulab.it/haze/misc/earth_2.png
http://mamedev.emulab.it/haze/misc/earth_3.png
http://mamedev.emulab.it/haze/misc/earth_4.png


The Cutscenes where you talk to people in the Link/Zelda 'twins' don't seem to work all the time, often they display nothing until you press a button? The one in the first room of the first area of Link never seems to play with the image I have.

Posted By: Robbbert

Re: SVN builds - new driver flood - 10/25/09 09:59 AM

Some more game.com...



Posted By: Christina

Re: SVN builds - new driver flood - 10/25/09 03:36 PM

Quite impressible!
Thank you!
Posted By: Haze

Re: SVN builds - new driver flood - 10/25/09 03:58 PM

CDi - more testing

Alice in Wonderland
Seems to have issues with the inputs? Tutorial tells you that you can use a button to enter doors / interact with things, but they don't seem to work

Also locks up when '4' is displayed during tutorial. If you start the game you can just move left/right around the world


http://mamedev.emulab.it/haze/misc/alice_1.png
http://mamedev.emulab.it/haze/misc/alice_2.png
http://mamedev.emulab.it/haze/misc/alice_3.png
http://mamedev.emulab.it/haze/misc/alice_4.png

Jokers Wild
Seems perfect.

http://mamedev.emulab.it/haze/misc/jokers_1.png
http://mamedev.emulab.it/haze/misc/jokers_2.png
http://mamedev.emulab.it/haze/misc/jokers_3.png
http://mamedev.emulab.it/haze/misc/jokers_4.png
http://mamedev.emulab.it/haze/misc/jokers_5.png
http://mamedev.emulab.it/haze/misc/jokers_6.png

Axis & Allies
runs, but appears to use an unsupported video format when drawing the title text animation (see shot 3)

Severe in-game lag when it changes between the mini-videos, leaving the entire ui unresponsive for several seconds at a time while you're trying to make decisions.

http://mamedev.emulab.it/haze/misc/axis_1.png
http://mamedev.emulab.it/haze/misc/axis_2.png
http://mamedev.emulab.it/haze/misc/axis_3.png
http://mamedev.emulab.it/haze/misc/axis_4.png
http://mamedev.emulab.it/haze/misc/axis_5.png
http://mamedev.emulab.it/haze/misc/axis_6.png
http://mamedev.emulab.it/haze/misc/axis_7.png
http://mamedev.emulab.it/haze/misc/axis_8.png
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/25/09 05:49 PM

Originally Posted By Haze
CDi - more testing

Alice in Wonderland
Seems to have issues with the inputs? Tutorial tells you that you can use a button to enter doors / interact with things, but they don't seem to work


I'll bet the third input bit that has a special meaning in graphics tablet operation actually functions as a third button. There's possibly even a fourth during normal gamepad operation. I'll see what I can do.

Originally Posted By Haze
Axis & Allies
runs, but appears to use an unsupported video format when drawing the title text animation (see shot 3)


Excellent, I've been trying to find a game with unsupported graphics modes, I'll see what that's trying to do. If you can find a game that actually uses RGB555 mode, too, that would be stellar.

Originally Posted By Haze
Burn Cycle works pretty well too, aside from some random graphic corruption, which appears to be self-correcting as it vanishes if you turn around again.


Probably the same issue that causes Myst's display to freak out during the transition when turning to the left or right, but to rectify itself once the transition is complete. I've been meaning to look at that, too.

If you could concentrate your testing on games that have graphical issues rather than lag or audio issues, that would be extremely helpful as well. Lemmings is currently unhappy with something to do with audio, as it only plays the first couple sectors' worth of background music before dying; I expect that the same thing that's causing this could be causing several of the audio issues you've described.
Posted By: Haze

Re: SVN builds - new driver flood - 10/25/09 05:58 PM

Family Games 2

Some of the minigames work, others have issues. This one 'Room Service' seems to have problems with moving up (only left/right/down work) There is also some graphical corruption as you can see.

http://mamedev.emulab.it/haze/misc/family2_1.png
http://mamedev.emulab.it/haze/misc/family2_2.png
http://mamedev.emulab.it/haze/misc/family2_3.png

Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/25/09 05:59 PM



Its time for mamma's coco...But she locked herself in.So you better find her in time
Posted By: Haze

Re: SVN builds - new driver flood - 10/25/09 08:39 PM

Hmm .. A CD-i game that doesn't suck! It actually has good art, nice animation and smooth scrolling!

http://mamedev.emulab.it/haze/misc/appren_1.png
http://mamedev.emulab.it/haze/misc/appren_2.png
http://mamedev.emulab.it/haze/misc/appren_3.png
http://mamedev.emulab.it/haze/misc/appren_4.png
http://mamedev.emulab.it/haze/misc/appren_5.png
http://mamedev.emulab.it/haze/misc/appren_6.png
http://mamedev.emulab.it/haze/misc/appren_7.png
http://mamedev.emulab.it/haze/misc/appren_8.png

The image I converted was a bit weird tho, and I get no music (and I think you need the digital board to get SFX because it just says SFX N/A in the menu)

Still.. that was a nice surprise
Posted By: Dr.Zer0

Re: SVN builds - new driver flood - 10/26/09 12:50 AM

Originally Posted By Haze


(and I think you need the digital board to get SFX because it just says SFX N/A in the menu)



in fact it requires the "Digital Video Card" for sound effects and various graphics enhancements...
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/26/09 03:42 AM

More Mac progress. And this is all real emulation - Basilisk II patches out the input, video, audio, and disk drivers with traps up to HLE native code, so they're doing a lot less than MESS is smile



Click here for full-size screenshot.
Posted By: Justin

Re: SVN builds - new driver flood - 10/26/09 04:51 AM

Hmm, MacMAME 0.28 apparently runs on 68k Macs....
Posted By: Phil Bennett

Re: SVN builds - new driver flood - 10/26/09 08:19 AM

Nice.
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 10/26/09 08:20 AM

Why the weird colours?
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/26/09 08:53 AM

Originally Posted By Justin
Hmm, MacMAME 0.28 apparently runs on 68k Macs....


we might end up with the first instance of MESS emulating itself... and on mac!
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/26/09 11:06 AM

In svn rev6190, I fixed the crash in cdtv. Now we can once more load CHDs to cdtv. (thanks Anna for providing a test case wink )

However, I think there is something wrong going on with the cdda (probably on the MAME side): the cdda can be set from the cdrom by using cdda_set_from_cdrom, but this function requires the cdrom to be a 'void' entry rather than a device (like cdroms are in MESS) and this was causing the error I fixed: the cdda.c code was not finding the correct cdda device when cdda_set_from_cdrom was called.
This is probably due to the fact that in MAME cdda_set_from_cdrom is called with a second parameter from SCSI elements and SCSI is not a device yet in MAME.

While the current code in MESS seems to work, there are other functions which expect a void parameter and receive a device one.
Hence, I fear there could be other problems (hopefully not affecting anything visible) which will only be fixed with a device-ification of SCSI!!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/26/09 01:01 PM

Originally Posted By Vas Crabb
Why the weird colours?


That's the default gamma for the Mac II. Seriously. I think you can change it in one of the control panels.

eta: that change is somewhere in the neighborhood of horrifically wrong. You should've just fixed cdda_from_cdrom.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/26/09 01:04 PM

Great fix, etabeta78 smile

SVN r6190
cdtv driver

Posted By: etabeta78

Re: SVN builds - new driver flood - 10/26/09 01:54 PM

Originally Posted By R. Belmont
eta: that change is somewhere in the neighborhood of horrifically wrong. You should've just fixed cdda_from_cdrom.


fixing cdda_from_cdrom requires changes in the way scsi works in drivers using that call (affecting among others gdrom.c and cubocd32.c). changes to scsi to pass a device in those two drivers might require a full device-ification of scsi, and hence touching the drivers using it (firebeat, ksys573 and dc among others)

I hope you realize it's not something I could have fixed in half an hour, like adding a tag in the interface. proper fix is in my (growing) todo list.

Given that I clearly documented the reason of the ghastly hack (so that the system will work in 0.135 but we know the fix is not good), I think we can live with it until I have some more spare time wink
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/26/09 02:26 PM

That's fair. I've deliberately avoided converting all of that to devices because MAME's irritating "startup order is not guaranteed" policy means we could easily end up with a situation where sometimes it can find the media and sometimes not.
Posted By: Duke

Re: SVN builds - new driver flood - 10/26/09 02:37 PM

Can't you just use device_delay_init() to guarantee that another device has been started?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/26/09 02:44 PM

Does that actually work? I vaguely recall someone tried to use it once and ended up quitting the scene ;-)
Posted By: judge

Re: SVN builds - new driver flood - 10/26/09 02:56 PM

If it doesn't you can always fix it :P
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 10/26/09 03:20 PM

Originally Posted By Just Desserts
I'll bet the third input bit that has a special meaning in graphics tablet operation actually functions as a third button. There's possibly even a fourth during normal gamepad operation. I'll see what I can do.

Nope, "button 3" on more recent controllers just presses 1 & 2 simultaneously smile There IS no button 3...
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 10/26/09 03:24 PM

Originally Posted By Haze
The image I converted was a bit weird tho, and I get no music (and I think you need the digital board to get SFX because it just says SFX N/A in the menu)

It's a CD-i Ready disc; without SFX is plays music from the CDDA audio tracks. To get SFX you need 1 MB of extra memory, nothing else.

The DV cartridge is one way of getting that, using a CD-i 605 another (it has more memory). JD, could you just add 1MB of RAM somewhere high (E80000 comes to mind)?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/26/09 03:45 PM

I could swear I already added an extra meg of memory at 0xE00000, is there any additional magic that needs to be fiddled with to make it recognize that a DVC is inserted?

I also need to hook up CD-DA audio support, I'll probably do that tonight.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/26/09 03:47 PM

Originally Posted By CD-i Fan
Nope, "button 3" on more recent controllers just presses 1 & 2 simultaneously smile There IS no button 3...


Maybe it's expecting me to press both buttons at the same time, then, and since it's unlikely that I can trigger both buttons in the same frame manually, it isn't working?
Posted By: Stiletto

Re: SVN builds - new driver flood - 10/26/09 04:15 PM

Amazing, Arbee. Excellent work.
Posted By: LuigiBlood

Re: SVN builds - new driver flood - 10/26/09 04:41 PM

Maybe it's about the Play, Pause, and Stop buttons from the remote controller ? I remember that i did pause Hotel Mario by pressing Pause.

EDIT :
http://www.retro-trader.com/images/june2003/DVC02205.JPG

EDIT 2 : Nevermind.
Posted By: Haze

Re: SVN builds - new driver flood - 10/26/09 09:05 PM

CDI Shots of the day

Magic Eraser
Take Qix, remove all the fun, all the gameplay, and all the rules, and you basically have Magic Eraser.

It looks like the background gfx should be 16bpp, maybe this is what you were looking for MG? (or something else is wrong.)

http://mamedev.emulab.it/haze/misc/magiceraser_1.png
http://mamedev.emulab.it/haze/misc/magiceraser_2.png
http://mamedev.emulab.it/haze/misc/magiceraser_3.png
http://mamedev.emulab.it/haze/misc/magiceraser_4.png

MegaMaze

This looked fun on YouTube, but just locks up in my build
http://mamedev.emulab.it/haze/misc/megamaze_1.png
http://mamedev.emulab.it/haze/misc/megamaze_2.png

Shaolin's Road

Seems like a well-enough done movie based game, nohing special, but more polished than most. It seemed playable up until the last shot, where I was hit for a second time, and instead of giving me the usual retry option it locked up

http://mamedev.emulab.it/haze/misc/shao_1.png
http://mamedev.emulab.it/haze/misc/shao_2.png
http://mamedev.emulab.it/haze/misc/shao_3.png
http://mamedev.emulab.it/haze/misc/shao_4.png
http://mamedev.emulab.it/haze/misc/shao_5.png
http://mamedev.emulab.it/haze/misc/shao_6.png
http://mamedev.emulab.it/haze/misc/shao_7.png
http://mamedev.emulab.it/haze/misc/shao_8.png
http://mamedev.emulab.it/haze/misc/shao_9.png
http://mamedev.emulab.it/haze/misc/shao_10.png
http://mamedev.emulab.it/haze/misc/shao_11.png
http://mamedev.emulab.it/haze/misc/shao_12.png
http://mamedev.emulab.it/haze/misc/shao_13.png
http://mamedev.emulab.it/haze/misc/shao_14.png
http://mamedev.emulab.it/haze/misc/shao_15.png
http://mamedev.emulab.it/haze/misc/shao_16.png
http://mamedev.emulab.it/haze/misc/shao_17.png

Video SpeedWay

The worst Super Monaco GP clone I've ever played. The menus are quite nice, but the game is awful. Seems to work fine, including the car that sounds like a broken lawnmower.

http://mamedev.emulab.it/haze/misc/vidspeed_1.png
http://mamedev.emulab.it/haze/misc/vidspeed_2.png
http://mamedev.emulab.it/haze/misc/vidspeed_3.png
http://mamedev.emulab.it/haze/misc/vidspeed_4.png
http://mamedev.emulab.it/haze/misc/vidspeed_5.png
http://mamedev.emulab.it/haze/misc/vidspeed_6.png
http://mamedev.emulab.it/haze/misc/vidspeed_7.png
http://mamedev.emulab.it/haze/misc/vidspeed_8.png
http://mamedev.emulab.it/haze/misc/vidspeed_9.png

Lords of The Rising Sun
Seems to work, strategy game, couldn't work out what to do on the map screen tho.

http://mamedev.emulab.it/haze/misc/lords_1.png
http://mamedev.emulab.it/haze/misc/lords_2.png
http://mamedev.emulab.it/haze/misc/lords_3.png
http://mamedev.emulab.it/haze/misc/lords_4.png
http://mamedev.emulab.it/haze/misc/lords_5.png
http://mamedev.emulab.it/haze/misc/lords_6.png
http://mamedev.emulab.it/haze/misc/lords_7.png

Asterix - worked up to the point of telling me that my disc was dirty, at which point it died.

http://mamedev.emulab.it/haze/misc/casterix_1.png
http://mamedev.emulab.it/haze/misc/casterix_2.png
http://mamedev.emulab.it/haze/misc/casterix_3.png
http://mamedev.emulab.it/haze/misc/casterix_4.png
http://mamedev.emulab.it/haze/misc/casterix_5.png
http://mamedev.emulab.it/haze/misc/casterix_6.png
http://mamedev.emulab.it/haze/misc/casterix_7.png
http://mamedev.emulab.it/haze/misc/casterix_8.png
http://mamedev.emulab.it/haze/misc/casterix_9.png

Kether - Locked up at the last shot

http://mamedev.emulab.it/haze/misc/kether_1.png
http://mamedev.emulab.it/haze/misc/kether_2.png
http://mamedev.emulab.it/haze/misc/kether_3.png
http://mamedev.emulab.it/haze/misc/kether_4.png
http://mamedev.emulab.it/haze/misc/kether_5.png
http://mamedev.emulab.it/haze/misc/kether_6.png
http://mamedev.emulab.it/haze/misc/kether_7.png
http://mamedev.emulab.it/haze/misc/kether_8.png
http://mamedev.emulab.it/haze/misc/kether_9.png

Domino - pointless, but works

http://mamedev.emulab.it/haze/misc/cdomino_1.png
http://mamedev.emulab.it/haze/misc/cdomino_2.png

Laser Lords
Seems to be based on the same engine as Alice in Wonderland, and also suffers from the 'can't do any actions on the surface' issue.

http://mamedev.emulab.it/haze/misc/laslords_1.png
http://mamedev.emulab.it/haze/misc/laslords_2.png
http://mamedev.emulab.it/haze/misc/laslords_3.png
http://mamedev.emulab.it/haze/misc/laslords_4.png
http://mamedev.emulab.it/haze/misc/laslords_5.png
http://mamedev.emulab.it/haze/misc/laslords_6.png
http://mamedev.emulab.it/haze/misc/laslords_7.png
http://mamedev.emulab.it/haze/misc/laslords_8.png

Voyeur (ADULT GAME)
point+click adventure? might have potential, but has lots of instant death situations, and instant death means.. show the credits and quit the game!.. The emulator doesn't return to the interface and boot the game again gracefully.

http://mamedev.emulab.it/haze/misc/voyuer_1.png
http://mamedev.emulab.it/haze/misc/voyuer_2.png
http://mamedev.emulab.it/haze/misc/voyuer_3.png
http://mamedev.emulab.it/haze/misc/voyuer_4.png
http://mamedev.emulab.it/haze/misc/voyuer_5.png
http://mamedev.emulab.it/haze/misc/voyuer_6.png
http://mamedev.emulab.it/haze/misc/voyuer_7.png

Vegas Girls (ADULT GAME)

Misalinged graphics, it's a PAL title, so maybe that's why? or it's something to do with the status bar, and timing, maybe i should check it with it disabled, although the code to me didn't look like it would change the timing. Also the graphics corrupt HEAVILY on the fruit machine when the reels spin (shots 1, 4-5, safe for work)

http://mamedev.emulab.it/haze/misc/vegasgl_1.png
http://mamedev.emulab.it/haze/misc/vegasgl_2.png
http://mamedev.emulab.it/haze/misc/vegasgl_3.png
http://mamedev.emulab.it/haze/misc/vegasgl_4.png
http://mamedev.emulab.it/haze/misc/vegasgl_5.png
http://mamedev.emulab.it/haze/misc/vegasgl_6.png
http://mamedev.emulab.it/haze/misc/vegasgl_7.png
http://mamedev.emulab.it/haze/misc/vegasgl_8.png

Strip Poker Pro (ADULT GAME)
seems to work, didn't bother to play for long

http://mamedev.emulab.it/haze/misc/strippp_1.png
http://mamedev.emulab.it/haze/misc/strippp_2.png
http://mamedev.emulab.it/haze/misc/strippp_3.png
http://mamedev.emulab.it/haze/misc/strippp_4.png
http://mamedev.emulab.it/haze/misc/strippp_5.png
http://mamedev.emulab.it/haze/misc/strippp_6.png

Strip Poker Live (ADULT GAME)
seems to work, didn't bother to play for long

http://mamedev.emulab.it/haze/misc/strippl_1.png
http://mamedev.emulab.it/haze/misc/strippl_2.png
http://mamedev.emulab.it/haze/misc/strippl_3.png
http://mamedev.emulab.it/haze/misc/strippl_4.png
http://mamedev.emulab.it/haze/misc/strippl_5.png

CD Shoot
the title probably indicates the best use for the CD after playing this game. Somebody please tell phillips that you don't create a mouse aiming game, then auto move the target to some random location on the screen when the thing you're trying to shoot appears, and attempt to track it, while still expecting the player to fine tune the movement. It just makes things unplayable. Appears to work tho.


http://mamedev.emulab.it/haze/misc/cdshoot_1.png
http://mamedev.emulab.it/haze/misc/cdshoot_2.png
http://mamedev.emulab.it/haze/misc/cdshoot_3.png

those are the ones I've tested today, I guess the only interesting one as far as MG is concerned will be the first one, although the Vegas Girls one might be of interest too. Shots just for people who are interested in seeing what they looked like.


At this point I genuinely feel sorry for anybody who paid $700 for one of these and $50 a game....

Posted By: Robbbert

Re: SVN builds - new driver flood - 10/26/09 10:13 PM

404 on the voyeur screenshots.
Posted By: Haze

Re: SVN builds - new driver flood - 10/26/09 10:26 PM

fixed, I can't spell ;-)

Sport Freaks actually gives a warning, which I'm assuming says it only works on European (PAL?) CD-Is and then quits

http://mamedev.emulab.it/haze/misc/sprtfrk_1.png

Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/26/09 10:54 PM

Can someone name off a few games that use CDDA audio? I have The Apprentice, but it's in some funky CUE/BIN format that Daemon Tools can't mount. frown
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/26/09 11:00 PM

Use CD-i Fan's direct converter - it's in that format that only CloneCD can read.
Posted By: Haze

Re: SVN builds - new driver flood - 10/26/09 11:01 PM

Originally Posted By Just Desserts
Can someone name off a few games that use CDDA audio? I have The Apprentice, but it's in some funky CUE/BIN format that Daemon Tools can't mount. frown


the one I found had a corrupt cuesheet, the first track was marked as Audio when it wasn't.

it had
FILE "The_Apprentice.img" BINARY
TRACK 1 AUDIO
INDEX 1 15:22:00
TRACK 2 AUDIO
INDEX 1 19:44:00
etc.

instead of

FILE "The_Apprentice.img" BINARY
TRACK 1 MODE2/2352
INDEX 1 00:00:00
TRACK 2 AUDIO
INDEX 1 15:22:00
TRACK 3 AUDIO
INDEX 1 19:44:00

I can't tell you if the 2nd is right for sure, but the initial cuesheet definitely doesn't work, so if you have the same image and cuesheet...

It's the only game I have with CDDA tho. I think A Wu posted another one a few pages back, Alien something?


Posted By: CD-i Fan

Re: SVN builds - new driver flood - 10/26/09 11:32 PM

Originally Posted By LuigiBlood
Maybe it's about the Play, Pause, and Stop buttons from the remote controller ? I remember that i did pause Hotel Mario by pressing Pause.

That is technically possible (the buttons can be read by a CD-i application), but I'd rate it "unlikely".
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 10/26/09 11:38 PM

Originally Posted By Haze
Sport Freaks actually gives a warning, which I'm assuming says it only works on European (PAL?) CD-Is and then quits

http://mamedev.emulab.it/haze/misc/sprtfrk_1.png

Yep, it says "This program works only on European CD-i players. Press a button on the remote to leave the program."

This title wasn't published by Philips who would never have allowed this cheap cop-out, but since all the text was in dutch anyway we didn't feel the need to bother. Yes, this was developed at the company I used to work for, I might have even written part of the code, can't remember exactly.
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 10/26/09 11:43 PM

Most games from SPC Vision / The Vision Factory have CDDA tracks, but only The Apprentice and Alien Gate play them from the game (only if you use the "NO SFX" setting, though for different reasons in each game). Rhese tracks were mostly intended for people to play on their audio players, which was the primary reason for the CD-i Ready format.

Don't know if either cue sheet for The Apprentice is technically right; CD-i Ready hides the CD-i data in the "pregap" of the first audio track because most audio players skip over that automatically. The first cue sheet is thus technically right, I think.

Edit: Both games always play their out-game music (e.g. intros, highscore, credits) as CDDA, regardless of the SFX setting
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 10/26/09 11:55 PM

Also, be aware that there are a LOT of bad images for CD-i Ready discs out there. If CD-i Emulator plays them, they're ok, otherwise I'd say "strongly suspect" smile

Some of them are okay in the CD-i track but bad in the audio part...
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/27/09 01:00 AM

Just added CDDA support. Enjoy.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/27/09 01:09 AM

Just backed out CDDA support. Don't enjoy. It was causing The Apprentice to play nice CDDA on its game intro screen, but at the same time it also didn't progress past the first frame of the game intro screen. frown
Posted By: Cpt. Pugwash

Re: SVN builds - new driver flood - 10/27/09 05:23 AM

Originally Posted By Haze
fixed, I can't spell ;-)

Sport Freaks actually gives a warning, which I'm assuming says it only works on European (PAL?) CD-Is and then quits

http://mamedev.emulab.it/haze/misc/sprtfrk_1.png



Correct, it's in Dutch. (I don't speak it, but it's similar to German).

The text means:

"This program solely runs on european CD-i players. Please press a button on the remote control to leave the program."
Posted By: Haze

Re: SVN builds - new driver flood - 10/27/09 07:32 AM

Originally Posted By Just Desserts
Just backed out CDDA support. Don't enjoy. It was causing The Apprentice to play nice CDDA on its game intro screen, but at the same time it also didn't progress past the first frame of the game intro screen. frown


Keep in mind that enabling CDDA breaks some Konami PSX based games too, IIRC Beat The Champ and the other one on that hardware. If you turn off 'Music' in test mode they play, if you enable music they just report never ending disc / load errors. It could be something broken somewhere else in MAMEs CD code when it comes to handling CDDA and it also requires other access to the disc?
Posted By: judge

Re: SVN builds - new driver flood - 10/27/09 07:41 AM

I wonder if regular CDDA playback still works in the PCE music player then.
Posted By: Haze

Re: SVN builds - new driver flood - 10/27/09 07:55 AM

Originally Posted By judge
I wonder if regular CDDA playback still works in the PCE music player then.


I think 'basic' cases are fine, it just seems to be cases where it wants to do something else as well.

I was looking over the video timing in the driver, and it seems a bit strange.

The LCD seems to be taken into account for the video timing, however, if it's not an actual feature of the video chip, as posts here seem to suggest, that doesn't make much sense to me.

Of course you can just hide the first 22 lines, as you did before, but if it's another device controlling the LCD then the processing that happens on scanline 22 in mcd212_perform_scan should really be happening in scanline 0, at the same time as the LCD output, but with the LCD output writing to a different screen bitmap.

However, moving the code around just breaks it completely, as the processing done at line 0 right now can't happen at the same time as the processing done at line 22, even if line 22 seems to just be some arbitrary line based on the LCD height.

There might be a logical reason for all this (ie the same video chip really draws the LCD), I'm just not really understanding it.
Posted By: Christina

Re: SVN builds - new driver flood - 10/27/09 09:37 AM

I saw the recent improvements in CDTV system. But in the game "Turrican" i have the output:
"CPU maincpu: PULSE_LINE can only be used for NMI and RESET lines"
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/27/09 09:45 AM

that's an unrelated problem, but thanks for pointing out.

I don't see exactly why that error occurs, but I will see if I can find it...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/27/09 09:47 AM

Originally Posted By etabeta78
that's an unrelated problem, but thanks for pointing out.

I don't see exactly why that error occurs, but I will see if I can find it...


You will get a PM in the next minutes. smile
Posted By: Haze

Re: SVN builds - new driver flood - 10/27/09 12:02 PM

Originally Posted By Haze
Originally Posted By judge
I wonder if regular CDDA playback still works in the PCE music player then.


I think 'basic' cases are fine, it just seems to be cases where it wants to do something else as well.

I was looking over the video timing in the driver, and it seems a bit strange.

The LCD seems to be taken into account for the video timing, however, if it's not an actual feature of the video chip, as posts here seem to suggest, that doesn't make much sense to me.

Of course you can just hide the first 22 lines, as you did before, but if it's another device controlling the LCD then the processing that happens on scanline 22 in mcd212_perform_scan should really be happening in scanline 0, at the same time as the LCD output, but with the LCD output writing to a different screen bitmap.

However, moving the code around just breaks it completely, as the processing done at line 0 right now can't happen at the same time as the processing done at line 22, even if line 22 seems to just be some arbitrary line based on the LCD height.

There might be a logical reason for all this (ie the same video chip really draws the LCD), I'm just not really understanding it.


It also strikes me that the current timer system could quite easily cancel out previous timers if commands are sent quickly, which may or may not be desired... Does MAME/MESS provide a way of seeing if a timer has cancelled one which was already running? I guess some commands could quite easily be being lost that way...
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/27/09 12:34 PM

Originally Posted By Haze
It also strikes me that the current timer system could quite easily cancel out previous timers if commands are sent quickly, which may or may not be desired... Does MAME/MESS provide a way of seeing if a timer has cancelled one which was already running? I guess some commands could quite easily be being lost that way...


It isn't either of the things you've proposed. I would suggest looking at an error log before theorizing further - the CD-i can't perform any other disc access commands while CDDA is playing, so that's a non issue. Also, the LCD handling is fine, there are 22 lines of blanking at the start of each frame. I'm just choosing to shove LCD drawing into those 22 lines and to set the visible area to show it rather than just handling ICA at line 0 and setting the visible area down further.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/27/09 01:57 PM

The Konami PSX games are broken because of a specific problem with the emulation of the 53cf96 status bits, it's not a general issue with CDDA.

Also, it's physically impossible for any drive to load data while CDDA is playing. MK3 on the PSX had CDDA music and famously stopped it to load every time you did a character morph special move.
Posted By: Haze

Re: SVN builds - new driver flood - 10/27/09 02:06 PM

Originally Posted By R. Belmont
The Konami PSX games are broken because of a specific problem with the emulation of the 53cf96 status bits, it's not a general issue with CDDA.

Also, it's physically impossible for any drive to load data while CDDA is playing. MK3 on the PSX had CDDA music and famously stopped it to load every time you did a character morph special move.


right, I was thinking more along the lines of it leaving the drive in a bad state, rather than just stopping the audio when something else was requested.
Posted By: Haze

Re: SVN builds - new driver flood - 10/27/09 02:09 PM

Originally Posted By Just Desserts
Originally Posted By Haze
It also strikes me that the current timer system could quite easily cancel out previous timers if commands are sent quickly, which may or may not be desired... Does MAME/MESS provide a way of seeing if a timer has cancelled one which was already running? I guess some commands could quite easily be being lost that way...


It isn't either of the things you've proposed. I would suggest looking at an error log before theorizing further - the CD-i can't perform any other disc access commands while CDDA is playing, so that's a non issue. Also, the LCD handling is fine, there are 22 lines of blanking at the start of each frame. I'm just choosing to shove LCD drawing into those 22 lines and to set the visible area to show it rather than just handling ICA at line 0 and setting the visible area down further.


Ok, that wasn't really obvious from the code.

As for the error logging, nothing stands out, the second theory was actually more related to the crashes that happen when you're repeating actions, or doing things quickly. Othello is a good one to crash if you try quickly flipping between the screens, if you take your time it's fine.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/27/09 02:33 PM

Originally Posted By Haze
As for the error logging, nothing stands out, the second theory was actually more related to the crashes that happen when you're repeating actions, or doing things quickly. Othello is a good one to crash if you try quickly flipping between the screens, if you take your time it's fine.


When you say "crash", you mean it locks up in MESS, but doesn't otherwise cause MESS to die, correct? Just so I know what to look for.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/27/09 04:52 PM

SVN r6195
gamecom driver
crosshair support


Posted By: Haze

Re: SVN builds - new driver flood - 10/27/09 06:58 PM

Originally Posted By Just Desserts
Originally Posted By Haze
As for the error logging, nothing stands out, the second theory was actually more related to the crashes that happen when you're repeating actions, or doing things quickly. Othello is a good one to crash if you try quickly flipping between the screens, if you take your time it's fine.


When you say "crash", you mean it locks up in MESS, but doesn't otherwise cause MESS to die, correct? Just so I know what to look for.


Lemmings and one other (I can't remember which, it only happened once and wasn't reproducable) actually crash the emulator, right back to the command line.

The others are just ingame lockups and usually occur when transitioning between scenes, or if something needs loading, even if that transition has worked previously.

and screenshots for today come from Dimo's Quest. This one works really well, although the controls seem a bit too sensitive (could be just like that tho, even on the lowest digital speed it seems to move 2 tiles if you hold the button down for more than a tap) It's actually a good game if you can get over that tho, I guess every platform needs to have 1 good developer, it's by the same guys as The Apprentice.

http://mamedev.emulab.it/haze/misc/dimos_1.png
http://mamedev.emulab.it/haze/misc/dimos_2.png
http://mamedev.emulab.it/haze/misc/dimos_3.png
http://mamedev.emulab.it/haze/misc/dimos_4.png
http://mamedev.emulab.it/haze/misc/dimos_5.png
http://mamedev.emulab.it/haze/misc/dimos_6.png
http://mamedev.emulab.it/haze/misc/dimos_7.png
http://mamedev.emulab.it/haze/misc/dimos_8.png
http://mamedev.emulab.it/haze/misc/dimos_9.png
http://mamedev.emulab.it/haze/misc/dimos_10.png
http://mamedev.emulab.it/haze/misc/dimos_11.png
http://mamedev.emulab.it/haze/misc/dimos_12.png
http://mamedev.emulab.it/haze/misc/dimos_13.png
Posted By: Robbbert

Re: SVN builds - new driver flood - 10/27/09 08:10 PM

Some I prepared earlier..

Phone book, Solitare, Robotron, Calendar

Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/27/09 08:20 PM

Very nice smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/27/09 10:06 PM

r6197 adds a -createuncomphd option to chdman. It works identically to -createhd but the resulting CHD is not compressed and not write protected. Would everyone who complained about this (mizapf, I'm looking at you) please test this before I throw it over to MAME?

Also, the RasterOps video board for the Mac is completely emulated in all modes now:

Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/28/09 05:37 AM

Hmmmm... http://mahlemiut.marpirc.net/x68k_netbsd.png (r6202)
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/28/09 05:41 AM

I assume that goes down in a blaze of fatalerror()s from the PMMU? smile Still, pretty cool.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/28/09 06:10 AM

Yes, but it does at least load it in, with help from a little SCC kludge, and display something which disappears too quickly when the first PMMU fatalerror appears smile

http://mahlemiut.marpirc.net/x68k_netbsd_booting.png
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/28/09 07:19 AM

This is the last frame before the fatalerror():
680x0 PMMU: Unhandled Table B mode 0 (addr_in 02800070)

Posted By: etabeta78

Re: SVN builds - new driver flood - 10/28/09 07:52 AM

Originally Posted By R. Belmont
That's fair. I've deliberately avoided converting all of that to devices because MAME's irritating "startup order is not guaranteed" policy means we could easily end up with a situation where sometimes it can find the media and sometimes not.


it turned out that CDDA code is indeed looking for the media before it can be found. for the time being, I see no better way to fix this than the current hack (i.e. by specifying in the interface which CDDA device should be used). I will take in any case a look at the scsi code, later...
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/28/09 08:39 AM

Originally Posted By Christina
I saw the recent improvements in CDTV system. But in the game "Turrican" i have the output:
"CPU maincpu: PULSE_LINE can only be used for NMI and RESET lines"


this has been properly fixed in svn 6203: the driver was checking for carts even if no cart could have been present.

now both Turrican and Turrican II (and probably other games as well) work fine.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 10/28/09 12:01 PM

SVN r6203
cdtv driver
Game: Turrican and Turrican II (chd) with great music (Chris Huelsbeck)



Posted By: Haze

Re: SVN builds - new driver flood - 10/28/09 12:03 PM

Originally Posted By Haze
Originally Posted By Just Desserts

When you say "crash", you mean it locks up in MESS, but doesn't otherwise cause MESS to die, correct? Just so I know what to look for.


Lemmings and one other (I can't remember which, it only happened once and wasn't reproducable) actually crash the emulator, right back to the command line.

The others are just ingame lockups and usually occur when transitioning between scenes, or if something needs loading, even if that transition has worked previously.


For referece Sargon Chess crashes the actual emulator quite often (exit to commandline), as does 3rd Degree.

Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/28/09 12:54 PM

Try recompiling with verbose logging enabled, and a log level of 5. I'm willing to bet that every game that's crashing is doing so when it tries to perform a DMA from main memory location 0x05140000 or thereabouts. That seems to be the standard "Something has gone horribly wrong with audio" freak-out that I've been trying to figure out.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/28/09 01:08 PM

Originally Posted By mahlemiut
This is the last frame before the fatalerror():
680x0 PMMU: Unhandled Table B mode 0 (addr_in 02800070)


Hmm. Mode 0 is invalid and would crash on hardware too. Usually it means something's gone wrong parsing the tables - can you point me to the boot floppy that does that? (It's obviously OK to post links to NetBSD ;-)
Posted By: AaronGiles

Re: SVN builds - new driver flood - 10/28/09 02:00 PM

Originally Posted By R. Belmont
That's fair. I've deliberately avoided converting all of that to devices because MAME's irritating "startup order is not guaranteed" policy means we could easily end up with a situation where sometimes it can find the media and sometimes not.

That's what dependencies are for. At startup, a device can query other devices it depends on and see if they are started. If they are not, it can call device_delay_init() and exit, meaning it will be re-queued to the end.
Posted By: etabeta78

Re: SVN builds - new driver flood - 10/28/09 02:03 PM

would you mind to offer an example of a device using this mechanism properly? thanks smile
Posted By: Duke

Re: SVN builds - new driver flood - 10/28/09 02:21 PM

See ldcore.c:

Code:
	/* ensure that our screen is started first */
	ld->screen = devtag_get_device(device->machine, config->screen);
	assert(ld->screen != NULL);
	if (!ld->screen->started)
	{
		device_delay_init(device);
		return;
	}


Posted By: etabeta78

Re: SVN builds - new driver flood - 10/28/09 02:34 PM

mmm... I never noticed the xxxx->started state... blush
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/28/09 03:38 PM

Incidentally, Haze, about Mega Maze or whatever it's called - it did work at one point, but it broke shortly after I started adding audio support. I might see if there's a way-old version of the driver from which I can glean some insight.
Posted By: Kale

Re: SVN builds - new driver flood - 10/28/09 11:48 PM

r6210 /src/mess/drivers/act.c: [ApricotF1]: Implemented some basic stuff

Looks fun. smile





Error code 29 means that the z80ctc failed, the only cpu there is a i8086...
Posted By: Just Desserts

Re: SVN builds - new driver flood - 10/28/09 11:50 PM

Originally Posted By Just Desserts
Incidentally, Haze, about Mega Maze or whatever it's called - it did work at one point, but it broke shortly after I started adding audio support. I might see if there's a way-old version of the driver from which I can glean some insight.


Yyyyyyeah, I'm just about out of ideas as to what Mega Maze wants that's so special, but whatever it is, I'll bet it's the same problems as Lemmings - it plays a short bit of audio, then it issues a Reset Mode 2 command, which implies that it was unhappy with something.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/29/09 01:26 AM

http://mahlemiut.marpirc.net/netbsd_install.zip (~1.1MB)

This is the disk image I created (blank 2HC disk image created with XM6, written from image file via rawrite (Human68k version under MESS, no less)). It's a standard NetBSD/x68k (v3.1, I think) installation disk. No way it'll be actually installable though, until the SCSI controller is emulated (missing ROM dump). NetBSD does not support SASI HDs.

Edit: Oh yeah, almost forgot, to allow it to progress past the initial logo, you have to kludge the SCC - just change line 593 from return scc_r(scc,1); to return 0x04;
Posted By: R. Belmont

Re: SVN builds - new driver flood - 10/30/09 04:03 AM

Good news/bad news on that. The good news is NetBSD is getting far enough into initialization to start wanting to take pagefaults. The bad news is the entire 68k core has to be rewritten from scratch to allow instructions to save their state in the middle and restart with no side effects after the page fault handler runs, and I am not that crazy :-)

ETA: There's a real hardware basis for this as well - the original 68000 was also unable to save instruction state in the middle and thus you can't have demand-paged memory on a 68000 system (the '010 and later are able to do it). There is only 1 known 68k software emulator that allows it and can boot Linux and NetBSD to userland, that being the core from the Atari ST emulator Aranym which is also used in recent WinUAE betas.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 10/30/09 12:07 PM

Oh, well. Maybe we can start getting this working. Works with SASI HDs apparently too. smile

Posted By: John Elliott

Re: SVN builds - new driver flood - 11/01/09 02:31 PM

Originally Posted By Kale
r6210 /src/mess/drivers/act.c: [ApricotF1]: Implemented some basic stuff

Error code 29 means that the z80ctc failed, the only cpu there is a i8086...


Yes, the F1 has an 8086 but a lot of the support chips are from the Z80 chipset. Meaning that, among other things, it does fake Z80 instruction fetches to acknowledge interrupts smile
Posted By: Kale

Re: SVN builds - new driver flood - 11/01/09 04:09 PM

Yes, puts on port $30 the string 0xed 0x4d, that's a RETI opcode in z80-ese. smile
It needs z80ctc & z80sio changes to support the i8086 hook-up anyway...
Posted By: Multipass

Re: SVN builds - new driver flood - 11/01/09 07:23 PM

Hello all, i have just one question, why psx driver not have support cd rom and chd? All drivers work of course?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/01/09 07:26 PM

Originally Posted By Multipass
Hello all, i have just one question, why psx driver not have support cd rom and chd? All drivers work of course?


MAME's support for Playstation-based arcade games is somewhat iffy. smf supposedly has a bunch of changes that make it more accurate, but nobody knows when those are going in, and I don't remember the last time any forward progress has been made in smf's Playstation research anyway. He's more than welcome to pop into the thread and prove me wrong, however. smile

Last I checked, MESS doesn't even boot the Playstation BIOS properly anymore anyway, so asking it to run any games is kind of putting the cart before the horse. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/01/09 07:31 PM

it was booting exe though at a point (the demo listed in the hash file)
Posted By: Justin

Re: SVN builds - new driver flood - 11/01/09 07:36 PM

It still runs some EXEs (e.g. Hitmen's Dentro).
Posted By: Multipass

Re: SVN builds - new driver flood - 11/01/09 07:36 PM

Originally Posted By Just Desserts
Originally Posted By Multipass
Hello all, i have just one question, why psx driver not have support cd rom and chd? All drivers work of course?


MAME's support for Playstation-based arcade games is somewhat iffy. smf supposedly has a bunch of changes that make it more accurate, but nobody knows when those are going in.

Last I checked, MESS doesn't even boot the Playstation BIOS properly anymore anyway, so asking it to run any games is kind of putting the cart before the horse. smile


"MAME's support for Playstation-based arcade games is somewhat iffy", you mean zn driver?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/01/09 07:49 PM

Yeah, the hook to make EXEs boot interrupts it before the BIOS intro starts so it's not showing is actually a sign that it's in good shape.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/01/09 08:08 PM

Originally Posted By Multipass
"MAME's support for Playstation-based arcade games is somewhat iffy", you mean zn driver?


Protip: There are many, many more games in MAME that are Playstation-based than just ZN.
Posted By: Multipass

Re: SVN builds - new driver flood - 11/01/09 08:08 PM

For example please? Someone work on this driver or no actually?
Posted By: Sune

Re: SVN builds - new driver flood - 11/01/09 08:15 PM

Originally Posted By Multipass
For example please?

Namco System 11 for example.
List of all games that use 'PSXCPU' in MAME: MAWS
Originally Posted By Multipass
Someone work on this driver or no actually?

http://www.bannister.org/forums/ubbthreads.php?ubb=showflat&Number=55861#Post55861
"smf supposedly has a bunch of changes that make it more accurate, but nobody knows when those are going in."

PSX is the new 2006.

Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/01/09 08:17 PM

Originally Posted By Multipass
Someone work on this driver or no actually?


Nobody is working on the driver right now.
Posted By: Multipass

Re: SVN builds - new driver flood - 11/01/09 08:49 PM

Originally Posted By Justin
It still runs some EXEs (e.g. Hitmen's Dentro).

please where i can find this exe please (if is authorized) ?
Posted By: Justin

Re: SVN builds - new driver flood - 11/01/09 09:06 PM

Originally Posted By Multipass
Originally Posted By Justin
It still runs some EXEs (e.g. Hitmen's Dentro).

please where i can find this exe please (if is authorized) ?


http://www.zophar.net/pdroms/psx/dentro.html

Hitmen's other demos are available at http://hitmen.c02.at/html/psx_releases.html (not all of them run though)

This demo runs well in MESS too: http://pouet.net/prod.php?which=3610

Posted By: Multipass

Re: SVN builds - new driver flood - 11/01/09 10:44 PM

More thanks for help , and links alll wink
Posted By: Heihachi_73

Re: SVN builds - new driver flood - 11/01/09 11:34 PM

I could 'only' come up with these 230-or-so games in MAME, along with a few more Namco games using a slightly more upbeat hardware.

Whether or not they are strictly PSX based is another story, I was just cheating a little by using MAWS. wink
Posted By: Kale

Re: SVN builds - new driver flood - 11/02/09 12:11 AM

r6242 /src/mess/drivers/act.c: [Apricot Xi]: Emulated basic video HW



Now this is odd: MC6845 sets 640x375, tilemap size is 80x25 and charset is 10x16...obviously, 80x10=800, so right-most columns gets cutted and I don't have any hell of an idea about it.
And...before you ask, no, start address is equal to 0 smirk
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/02/09 01:01 AM

Heihachi: a MAWS search on PSXCPU is authoritative if you filter out the BIOS sets.
Posted By: Sune

Re: SVN builds - new driver flood - 11/02/09 02:36 PM

The Mac Plus driver broke with the latest mac driver update, it deadmacs during POST.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/02/09 03:09 PM

Originally Posted By Sune
The Mac Plus driver broke with the latest mac driver update, it deadmacs during POST.


I confirm it.
Posted By: Sune

Re: SVN builds - new driver flood - 11/02/09 06:15 PM

wow. macplus works again in latest svn.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/02/09 06:40 PM

Originally Posted By Sune
wow. macplus works again in latest svn.


A very fast fix. smile
Thank you, Arbee.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/02/09 07:38 PM

SVN r6261
gamegeaj driver
Game : Ayrton Senna's Super Monaco GP II (J)

1st screenshot without fix, 2nd screenshot with fix




Posted By: Haze

Re: SVN builds - new driver flood - 11/02/09 08:16 PM

Originally Posted By Anna Wu
SVN r6261
gamegeaj driver
Game : Ayrton Senna's Super Monaco GP II (J)

1st screenshot without fix, 2nd screenshot with fix






c'mon, this is an ugly hack.

The LCD persistence isn't a feature of the video chip, and shouldn't be emulated in the driver, it's misleading and wrong. It's a feature of the output screen device and should be emulated as an option in the core like the scanlines etc. If you hooked a gamegear up to a normal monitor somehow you wouldn't see it.

There are plenty of arcade games which flicker graphics to display them as transparent / half-bright (neogeo life-bars come to mind) but it's not up to the driver to simulate that behavior just because the flicker is more obvious on a PC monitor.


Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/02/09 08:31 PM

Yeah, this should happen in the core if the screen type is set to LCD, and there should be some way for drivers to "opt out" as well. Sega handhelds were notoriously smeary (playing Sonic 2 on a Nomad is a real experience for that reason), but others aren't so we don't want to pin drivers that don't have the effect with it.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/02/09 08:37 PM

Originally Posted By R. Belmont
Yeah, this should happen in the core if the screen type is set to LCD, and there should be some way for drivers to "opt out" as well. Sega handhelds were notoriously smeary (playing Sonic 2 on a Nomad is a real experience for that reason), but others aren't so we don't want to pin drivers that don't have the effect with it.


Either you or Haze are more than welcome to fix it. In the meantime, I'm off. The level of ingratitude for trying to slog through a few old bugs is making my head spin.
Posted By: Haze

Re: SVN builds - new driver flood - 11/02/09 08:39 PM

Originally Posted By R. Belmont
Yeah, this should happen in the core if the screen type is set to LCD, and there should be some way for drivers to "opt out" as well. Sega handhelds were notoriously smeary (playing Sonic 2 on a Nomad is a real experience for that reason), but others aren't so we don't want to pin drivers that don't have the effect with it.


Yep, and that way all the systems that need it benefit from it.

It's a common technique used on handhelds to overcome sprite limits and create transparent effects, so having it in the core means that any system that needs it can use it, having it in the drivers helps nothing in the longrun.

Is there any need for all the drama just because I disagree with something ?!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/02/09 08:40 PM

Like it or not, you two are the drama king and queen of emulation (you can fight on IRC over who's who, I won't allow it here).
Posted By: Haze

Re: SVN builds - new driver flood - 11/02/09 08:42 PM

Originally Posted By Just Desserts
Originally Posted By R. Belmont
Yeah, this should happen in the core if the screen type is set to LCD, and there should be some way for drivers to "opt out" as well. Sega handhelds were notoriously smeary (playing Sonic 2 on a Nomad is a real experience for that reason), but others aren't so we don't want to pin drivers that don't have the effect with it.


Either you or Haze are more than welcome to fix it. In the meantime, I'm off. The level of ingratitude for trying to slog through a few old bugs is making my head spin.


Well it's the same in MAME really, there are lots of old bugs which can be fixed with a hack, as Mamesick often does, but doing it that way often doesn't really bring us closer to a real solution and can bloat the code and get in the way of figuring things out properly.

Here the path is more obvious tho, although it needs somebody who does core / video output to handle it, rather than the people who work on the drivers. (Just as fixing the bugs Mamesick hacks need somebody skilled / capable of testing on the hardware to fix them properly rather than attempting to by simply reverse engineering of unclear/ambiguous behavior)
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/02/09 08:52 PM

What ever happened to hacks being acceptable as an interim solution on the way to a proper one, as long as they're marked as such in the source code?

I'd be more than happy to mark the code as a hack with some sort of a comment, but removing it entirely in favor of this non-existent core LCD persistence handling seems pointless.

On the topic of adding core support, the actual method of handling persistence is debatable. The Game Gear driver uses a flat 50/50 blend right now, but I expect other LCDs might have more or less persistence, so a two-frame average wouldn't work for those. Then, once it's supported by the core, what's the standard for determining the persistence value? Tweak it until it looks right? Write trojan code and use a high-quality camera to measure the persistence empirically?

Then, how do drivers specify persistence? Does it become a new MDRV parameter, MDRV_LCD_PERSISTENCE?

At any rate, I agree that it should eventually be moved into the core, but for heaven's sake, the bug has been there for 8 years, it's one of MESS's most long-standing bugs, and this is at least better than nothing.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/02/09 08:53 PM

I don't think this rises to the level of awfulness that Mamesick sometimes likes to perpetrate. At least the results are correct and pretty good looking, which I can't say for several Mamesick specials.

That said, this does clearly belong in the core as an optional attribute of screens that are type LCD.

ETA: Regarding tuning, I'm thinking you pass a struct that specifies how many frames to involve in the persistence and how much each contributes to the final image. It could be fiddly to tune, but you need that power.

Also, doing it in the core opens up the eventual possibility of having a pixel shader do it for you, which would be an obvious performance win.
Posted By: Haze

Re: SVN builds - new driver flood - 11/02/09 09:25 PM

Originally Posted By R. Belmont
I don't think this rises to the level of awfulness that Mamesick sometimes likes to perpetrate. At least the results are correct and pretty good looking, which I can't say for several Mamesick specials.

That said, this does clearly belong in the core as an optional attribute of screens that are type LCD.

ETA: Regarding tuning, I'm thinking you pass a struct that specifies how many frames to involve in the persistence and how much each contributes to the final image. It could be fiddly to tune, but you need that power.

Also, doing it in the core opens up the eventual possibility of having a pixel shader do it for you, which would be an obvious performance win.


There is a certain irony in the whole situation, that being the very first thing I ever did with MAME was release a couple of hacked builds which simulated an LCD screen for the output, by combining various amounts of the previous 2 screens. Of course, not having a clue how anything worked back then I destroyed all the normal output in the process, so it wasn't optional, but it was all good fun ;-) I mainly did it at the time as an example showing that it could be done and looked good, because none of the handheld emus had such a feature, and it was annoying me.

The pre-Aaron MAME D3D code actually had a 'feedback' option which could be used to give a similar effect, but it was ripped out along with everything else when Aaron added his own D3D implementation.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/02/09 09:38 PM

Completely off-topic, but the topic's long enough as it is - how do I manually return data from an SMH_BANK handler?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/02/09 09:45 PM

Huh? I'm pretty sure asking that means you're doing something wrong. Explain smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/02/09 09:49 PM

True, it is probably wrong. I just don't quite understand in what way overlapping AM_RANGE macros override each other, so that's all I can go with right now.

Line 79, sms.c:

AM_RANGE(0x8000, 0x9FFF) AM_READWRITE(SMH_BANK(4), sms_cartram_w) /* ROM bank / on-cart RAM */

This will not work for Game Gear games that use the 93C46 EEPROM. The EEPROM lines are hooked up at 0x8000, and direct EEPROM access is also available from 0x8008 through 0x8088. I was considering making an sms_cartram_r, and optionally either reading from ROM or reading from EEPROM depending on both the address and whether or not it's a 93C46-using game.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/02/09 09:59 PM

Ahh. Since the driver sets up what each bank means and handles the switching you should be able to pull that data by other means. You may end up not even using a core bank then.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/02/09 10:11 PM

Originally Posted By R. Belmont
Ahh. Since the driver sets up what each bank means and handles the switching you should be able to pull that data by other means. You may end up not even using a core bank then.


Ggghhhh. Have you seen sms_mapper_w? It reads like stereo instructions. That's kind of lame that whatever it is that SMH_BANK does can't be manually invoked.

That said, I think I have a lead, so... yeah.

Edit: Heck with it. I can't, for the life of me, figure out what it wants. Someone else can hook up the 93C46. It doesn't help that Meka's source code is harder to read than Greek.
Posted By: John Elliott

Re: SVN builds - new driver flood - 11/02/09 11:35 PM

Originally Posted By Kale
r6242 /src/mess/drivers/act.c: [Apricot Xi]: Emulated basic video HW

Now this is odd: MC6845 sets 640x375, tilemap size is 80x25 and charset is 10x16...obviously, 80x10=800, so right-most columns gets cutted and I don't have any hell of an idea about it.
And...before you ask, no, start address is equal to 0 smirk


Well, the size on a real Xi is 800x400 (80x10=800, 25x16=400) so I don't know where the 640x375 comes from. Is something assuming a character cell is 8 pixels wide and 15 high?
Posted By: Kale

Re: SVN builds - new driver flood - 11/02/09 11:51 PM

640x375 is the calculated MAME/MESS resolution from the MC6845 video core. I highly suspect that's wrong and you're right though, it's not even the first time that core betrays me...
Posted By: enik

Re: SVN builds - new driver flood - 11/03/09 03:43 AM

JD, with the code changed on SVN rev. 6261, variable "smsvdp->prev_bitmap_saved" is not necessary anymore. I was working on adding view modes for 3-D games, then updated the patch, attached it to bug 70 and included the changes to remove that variable. Please take a look on the patch.
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/03/09 06:10 AM

also, I think I left a smsvdp->tmpbitmap which was not necessary in the struct. feel free to remove it as well.


@JD: please move debug feature in CDI to CONFSETTING in place of DIPSETTING... dipswitches should only be the ones physically present on the hardware wink
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/03/09 12:59 PM

Originally Posted By etabeta78
@JD: please move debug feature in CDI to CONFSETTING in place of DIPSETTING... dipswitches should only be the ones physically present on the hardware wink


Will do! Looks like I'm going to be out of work for another day due to a stomach bug, so yeah. frown
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/03/09 02:38 PM

Is it possible to change a CPU's clock speed at runtime in MESS?

ETA: cputag_set_clock
Posted By: Carlos

Re: SVN builds - new driver flood - 11/03/09 10:19 PM

Probably it is not the correct thread, but I have fixed some of the problems in the GBC palette, this resolves some problem reported on bugzilla and make the graphic more clean.

diff:
http://www.mediafire.com/?zzjdjlh4wml

Wario Land 3 (World)



Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/03/09 11:43 PM

Speaking of graphics fixes...

SVN r6272
apple2e driver

I fixed the lo-res graphics issue described in Bugzilla #1842:

Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/04/09 12:30 AM

Oh, nice. I was going to look at that sometime but it kept slipping my attention smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/04/09 02:14 AM

SVN r6277
Sony PocketStation driver - in progress

Pocket Station seems to have something to say:


Posted By: Stiletto

Re: SVN builds - new driver flood - 11/04/09 02:16 AM

it appears to love you, JD... laugh
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/04/09 05:56 AM

SVN r6275

I can not compile wimgtool because I get following error:

Compiling src/mess/tools/imgtool/windows/wmain.c...
Compiling src/mess/tools/imgtool/windows/wimgtool.c...
Compiling src/mess/tools/imgtool/windows/attrdlg.c...
Compiling src/mess/tools/imgtool/windows/assoc.c...
cc1.exe: warnings being treated as errors
src/mess/tools/imgtool/windows/assoc.c: In function 'win_association_exists':
src/mess/tools/imgtool/windows/assoc.c:55: warning: implicit declaration of func
tion 'ARRAY_LENGTH'
mingw32-make: *** [obj/windows/mess/mess/tools/imgtool/windows/assoc.o] Error 1


So obj/windows/mess/mess/tools/imgtool/windows/assoc.o can not be created and is missing.
Because of this error, I can not compile other tools like chdman etc. too.

Posted By: JoJo

Re: SVN builds - new driver flood - 11/04/09 07:52 AM

That's probably my fault - Since I just build SDLMESS it slipped from my checks. I'm at work now, I'll give it a look later...
Posted By: Micko

Re: SVN builds - new driver flood - 11/04/09 07:53 AM

Have done fix already. Just take new code from SVN.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/04/09 08:40 AM

Thanks for the fix, Micko smile
Posted By: JoJo

Re: SVN builds - new driver flood - 11/04/09 11:47 AM

Originally Posted By Micko
Have done fix already. Just take new code from SVN.


Did you try including just "osdcomm.h" rather than "mame.h"?
Posted By: Robbbert

Re: SVN builds - new driver flood - 11/04/09 01:29 PM

I've noticed that GBA driver is no longer working.. at least for the roms i tried.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/04/09 02:19 PM

Originally Posted By robbbert
I've noticed that GBA driver is no longer working.. at least for the roms i tried.


I confirm it.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/04/09 02:34 PM

Heh. When did it break? Nobody's touched it in ages AFAIK.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/04/09 02:44 PM

I am not a expert.

Maybe the reason can be SVN r6243 or SVN r6181 ?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/04/09 02:52 PM

It was 6243. ACCESSING_ enforces 16-bit accesses in that case and the code was written to allow both 16 and 32-bit wide access (most games use a mix).
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/04/09 02:53 PM

Originally Posted By R. Belmont
It was 6243. ACCESSING_ enforces 16-bit accesses in that case and the code was written to allow both 16 and 32-bit wide access (most games use a mix).


I'll fix it this evening. Give me a specific game that it breaks, since the BIOS was fine the last time I checked.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/04/09 02:55 PM

Originally Posted By Just Desserts
Originally Posted By R. Belmont
It was 6243. ACCESSING_ enforces 16-bit accesses in that case and the code was written to allow both 16 and 32-bit wide access (most games use a mix).


I'll fix it this evening. Give me a specific game that it breaks, since the BIOS was fine the last time I checked.


The bios is ok but the driver can not load cartridge. Try as example " Super Mario Advance ".
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/04/09 03:06 PM

SVN r6288
The gba driver is now working again.
Thanks for the fix, Arbee. smile
Posted By: Robbbert

Re: SVN builds - new driver flood - 11/04/09 08:37 PM

All good now, now I can play Chip's Challenge again.. laugh

Thanks R.B.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/04/09 10:03 PM

SVN r6294
x1turbo40 driver
Game : Hyper Olympic 84 (.2d)



Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/04/09 10:13 PM

SVN r6298: Pocket Station driver (preliminary)
BIOS

It's now possible to adjust the date and time. The machine hangs if you press the Action button after setting the date and time, not sure why.

Posted By: Kale

Re: SVN builds - new driver flood - 11/04/09 10:44 PM

r6299 /src/mess/drivers/x1.c: [X1]: Heavily improved the kanji reading, this fixes most games that uses that feature

http://mamedev.emulab.it/kale/?p=819

afaik, just Asteka II and Hyper Olympics '84 doesn't work yet, the latter uses a different kanji reading method. And yes, this makes MESS to be the first X1 emulator to have this feature correctly implemented. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/04/09 11:02 PM

I love that someone in Italy is first with correct kanji hookup in an emulator for a JDM-only computer. Who was it with the theory that people only want to work on exotic-to-them hardware again? smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/05/09 04:37 AM

For anyone who hates the long bootup time of the CD-i driver, I've added savestate support in SVN revision 6302. Save a savestate at the main menu, and never have to sit through the 15-second bootup process again. Or until the savestate format changes again.
Posted By: Darkstar

Re: SVN builds - new driver flood - 11/05/09 05:45 AM

What exactly does the Kanji hookup do? Convert kanji to reading? Or is it used as IME that converts reading to kanji?

BTW while we're at the topic of japanese computers: Anyone know a decent place in Kyoto where I could get some of those old machines? I found lots of used game consoles so far but no computers... Or would I have to go to Akihabara for that?

-Darkstar
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/05/09 06:06 AM

SVN r6301
Sony Pocketstation driver
Game : Maze, Tetris (.gme)



Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/05/09 06:42 AM

Originally Posted By Darkstar
What exactly does the Kanji hookup do? Convert kanji to reading? Or is it used as IME that converts reading to kanji?

BTW while we're at the topic of japanese computers: Anyone know a decent place in Kyoto where I could get some of those old machines? I found lots of used game consoles so far but no computers... Or would I have to go to Akihabara for that?

-Darkstar


Maybe you find some info here.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/05/09 07:36 AM

Originally Posted By Kale
r6299 /src/mess/drivers/x1.c: [X1]: Heavily improved the kanji reading, this fixes most games that uses that feature

http://mamedev.emulab.it/kale/?p=819

afaik, just Asteka II and Hyper Olympics '84 doesn't work yet, the latter uses a different kanji reading method. And yes, this makes MESS to be the first X1 emulator to have this feature correctly implemented. smile


In other words, MESS is the first X1 emulator which use real rom dumps ?
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/05/09 07:49 AM

not only: even if you try to use the correct dump in other emulators, they would not be able to read the kanji from it in the way the games expect them (i.e. in the way a real x1 would read them)
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/05/09 07:56 AM

Understand, thanks etabeta78. smile
Posted By: Kale

Re: SVN builds - new driver flood - 11/05/09 05:03 PM

Originally Posted By etabeta78
not only: even if you try to use the correct dump in other emulators, they would not be able to read the kanji from it in the way the games expect them (i.e. in the way a real x1 would read them)


It doesn't work even if you use the fake kanji rom that they provides, granted that I've did everything correct (i.e. I've didn't bothered much)...
A small list of games that uses the kanji ROM are:
-Asteka II
-Gaia
-GaiFlame
-Gall Force
-Gambler Jiko Chuushinha
-Hyper Olympics '84
-Lost Armaggedon
-Maidum
-Maison Ikkoku - Omoideno Photograph
-M.U.L.E.

Maybe now I should put the student hat and try to implement .d88 writes...it's needed for later games like Sorcerian, King's Knight, American Truck, Hydlide floppy version, Legend of Kage... (And probably tells me what exactly is still needed for the z80dma to work)
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/06/09 01:36 PM

Originally Posted By Anna Wu
SVN r6301
Sony Pocketstation driver
Game : Maze, Tetris (.gme)


@JustDessert: mmm... Maze and any other game from Pocket MuuMuu I tried freeze right after title screen. do you have any idea about the reason? I could find some time to investigate it in the weekend, if you're not already working on it
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/06/09 01:42 PM

Originally Posted By etabeta78
@JustDessert: mmm... Maze and any other game from Pocket MuuMuu I tried freeze right after title screen. do you have any idea about the reason? I could find some time to investigate it in the weekend, if you're not already working on it


With the old driver, that was certainly the case. With the new driver, not so much. Some games have input issues, but they shouldn't just freeze. I just tried Maze, for instance, and it gets in-game, but there are input issues.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/06/09 01:53 PM

Originally Posted By Just Desserts
Originally Posted By etabeta78
@JustDessert: mmm... Maze and any other game from Pocket MuuMuu I tried freeze right after title screen. do you have any idea about the reason? I could find some time to investigate it in the weekend, if you're not already working on it


With the old driver, that was certainly the case. With the new driver, not so much. Some games have input issues, but they shouldn't just freeze. I just tried Maze, for instance, and it gets in-game, but there are input issues.


Yes, also with some other games like BlockOut. Cars seems to work.
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/06/09 02:04 PM

I stand corrected... it seems I started the older exe in place of the newer one. I can confirm those games work in the latest svn

well, I will maybe take a look to the support for .bin files (so that we can start the few remaining games we currently don't load properly) or to something else in MESS

thanks for the answer
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/06/09 02:08 PM

Originally Posted By etabeta78
I stand corrected... it seems I started the older exe in place of the newer one. I can confirm those games work in the latest svn

well, I will maybe take a look to the support for .bin files (so that we can start the few remaining games we currently don't load properly) or to something else in MESS

thanks for the answer


Good smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/06/09 02:51 PM

Originally Posted By etabeta78
well, I will maybe take a look to the support for .bin files (so that we can start the few remaining games we currently don't load properly) or to something else in MESS


If you'd like to, I won't stop you, but I was planning on continuing to work on the driver this evening and this weekend. smile
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/06/09 03:06 PM

I'll leave it to you, then. it's not like there is nothing else to work on wink
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/06/09 03:50 PM

Incidentally, any word on when 0.135 might get assembled? It'd be nice to put in a little extra polish to go with all the already awesome improvements and new systems smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/06/09 07:17 PM

SVN r6315
gbcolor driver
Game : Pokemon Pinball (U)

Thanks for the fixes, Carlos. smile




Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/06/09 10:44 PM

Originally Posted By etabeta78
I'll leave it to you, then.


I had a look at it. It looks evil. You can have it. smile
Posted By: Justin

Re: SVN builds - new driver flood - 11/07/09 02:15 AM

Originally Posted By Anna Wu
Thanks for the fixes, Carlos. smile




Bravo, I filed the bug on that one back in 2002 smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/07/09 03:47 PM

SVN r6322
Compile issue

H:\SVN\Bin\mess>mingw32-make TARGET=mess
Compiling src/mess/formats/ti85_ser.c...
cc1.exe: warnings being treated as errors
src/mess/formats/ti85_ser.c: In function 'device_load_ti85serial':
src/mess/formats/ti85_ser.c:1320: warning: passing argument 5 of 'ti85_convert_f
ile_data_to_serial_stream' discards qualifiers from pointer target type
mingw32-make: *** [obj/windows/mess/mess/formats/ti85_ser.o] Error 1

Edit: Solved. SVN\Bin\mess\src\mess\formats\ti85_ser.x before deleted.
Posted By: judge

Re: SVN builds - new driver flood - 11/07/09 04:12 PM

It's all good now, right?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/07/09 04:17 PM

Originally Posted By judge
It's all good now, right?


Yes, now it is ok. smile

SVN\Bin\mess\obj\windows\mess\mess\formats\ti85_ser.o was created.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/08/09 05:29 AM

Here's a picture worth about 3 dozen words.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/08/09 07:56 AM

maciicx

Excellent to see, the monitor settings can be saved now on uncompressed chd. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/08/09 01:48 PM

Yeah, the mouse sensitivity is the only setting that doesn't save right now and I'm not sure why. (That's PRAM related though, not disk).
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/09/09 08:42 AM

SVN r6343
intvkbd driver
Game : Frogger (cartridge)



PS: In the index, I see the tape support (similar to the Aquarius) which is not available in the intvkbd driver.
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/09/09 09:16 AM

what do you mean about the cassette? the keyboard add-on had a cassette port and hence MESS should support this eventually. I think it allowed the user to attach a tape recorder to save programs on tape.

it's in the list of things which should be added at the top of the driver.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/09/09 09:20 AM

Originally Posted By etabeta78
what do you mean about the cassette? the keyboard add-on had a cassette port and hence MESS should support this eventually. I think it allowed the user to attach a tape recorder to save programs on tape.

it's in the list of things which should be added at the top of the driver.


In the MESSUI the cassette port is not available. Maybe I should test the driver on command line.

Posted By: etabeta78

Re: SVN builds - new driver flood - 11/09/09 09:22 AM

it's not supported AT ALL at the moment... where did you read it is supported?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/09/09 09:26 AM

Originally Posted By etabeta78
it's not supported AT ALL at the moment... where did you read it is supported?


The index of the intvkbd (bios) show me.

Posted By: etabeta78

Re: SVN builds - new driver flood - 11/09/09 09:30 AM

ah ok, it appears there because it is an option the BIOS offers and we do not modify the BIOS (why should we??)

but of course it does not appear in the MESSUI because MESS does not support it... cassette support is listed both in the source and in sysinfo among the things that are not supported yet.

Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/09/09 09:33 AM

Originally Posted By etabeta78
ah ok, it appears there because it is an option the BIOS offers and we do not modify the BIOS (why should we??)

but of course it does not appear in the MESSUI because MESS does not support it... cassette support is listed both in the source and in sysinfo among the things that are not supported yet.



OK smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/10/09 05:39 AM

Regarding 6343, the reason PULSE_LINE is not allowed by the core is because no real hardware works that way. If you need that behavior, it's usually because you actually need to hook up the IRQ acknowledge register.
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/10/09 08:18 AM

amiga action replay cart would need a similar fix, if there is any amiga expert around...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/11/09 05:30 AM

SVN r6354

FM-Towns Marty driver (still in progress)
Logo



FM-Towns Car Marty driver (still in progress)
Logo



FM-Towns driver (still in progress)
Trying to load a disk




Posted By: mahlemiut

Re: SVN builds - new driver flood - 11/11/09 05:53 AM

It'll work even better when the i386 fixes are added (although not quite there, you can get to the A: prompt with some config.sys/autoexec.bat editing). Hmmm... maybe I ought to commit those i386 fixes, since they'll be in u1 anyway. smile
Posted By: Justin

Re: SVN builds - new driver flood - 11/11/09 06:49 AM

Yes, you ought to smile
Posted By: mahlemiut

Re: SVN builds - new driver flood - 11/11/09 07:55 AM

Done. smile

And dies with an invalid opcode not long after. But hey, it's progress. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/11/09 09:40 AM

Barry, can you post the content of your config.sys, please ?
It seems, something is wrong in my config.sys.

Posted By: Kale

Re: SVN builds - new driver flood - 11/11/09 09:44 AM

Originally Posted By mahlemiut

And dies with an invalid opcode not long after. But hey, it's progress. smile


Which one?
Posted By: mahlemiut

Re: SVN builds - new driver flood - 11/11/09 11:19 AM

It says "Invalid Opcode 0E". It calls a function that, for some reason, gets overwritten. I presume the function either shouldn't be called, or something is going nuts somewhere.
Posted By: Kale

Re: SVN builds - new driver flood - 11/11/09 01:35 PM

Probably it's some grouped multi-byte thing (i.e. it's not opcode 0x0e since that's push cs), i386 core goes nuts with unimplemented opcodes logging...
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/11/09 03:37 PM

Here a older DOS version. I removed all drivers in the config.sys/autoexec.bat

It seems, I not get the opcode error message with this version or I was not waiting long enough. smile



PS: The text message is from the autoexec.bat because the keyboard input is not supported yet.


Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/11/09 09:00 PM

SVN r6361
vp111, vip driver

Get a crash if I start the emulation.
Posted By: Darkstar

Re: SVN builds - new driver flood - 11/12/09 07:42 AM

Originally Posted By Anna Wu
Barry, can you post the content of your config.sys, please ?
It seems, something is wrong in my config.sys.


It tells you that it can't load SMARTDRIVE because HIMEM.SYS is not installed (the HMA message basically says the same: UMBs unavailable so the "dos=high" line is ignored and DOS is loaded into lowmem.)

...man how I miss those times ;-)

-Darkstar
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/12/09 08:04 AM

Thank you for the translation, Darkstar. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/12/09 08:07 AM

Originally Posted By Anna Wu
SVN r6361
vp111, vip driver

Get a crash if I start the emulation.


I have to say " sorry ". blush
Delete some related files like rca.x/vip.x etc. (src/obj), make update and compile it again.

Can load now vip/vp111 driver without a crash.

PS: Maybe not the best solution and a clean build is necessary again but I will do later. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/12/09 08:41 AM

SVN r6350
Victor/Interact/Hector driver

Victor



Interact



Hector1



Hec2hrp



Hec2mx80



PS: I ask Daniel Coulom for a tool to convert .k7 to .wav format (tape). For the opposite way exist a tool.


Posted By: etabeta78

Re: SVN builds - new driver flood - 11/12/09 09:38 AM

probably the k7 format is the same as the Thomson one, so we could add support for k7 directly.

also, imho, the keyboard emulation is currently an hack (it uses pc at keyboard shocked ), but it's definitely better than nothing wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/12/09 10:32 AM

Originally Posted By etabeta78
probably the k7 format is the same as the Thomson one, so we could add support for k7 directly.

also, imho, the keyboard emulation is currently an hack (it uses pc at keyboard shocked ), but it's definitely better than nothing wink

A .k7 support will be great. smile

PS: The Exl100 driver (not working yet) also use the .k7 format.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/12/09 06:07 PM

Daniel was so friendly to contact yo_fr, to ask for help.

The .k7 format structure between the Thomson and the Hector computer is very different.
Posted By: etabeta78

Re: SVN builds - new driver flood - 11/12/09 07:44 PM

oh, thanks. then it will require some more work.

it will be implemented eventually smile
Posted By: Robbbert

Re: SVN builds - new driver flood - 11/15/09 01:46 AM

In the debug build, all Apple2 and higher (apple//c, apple//e, the GS, and the Macs, are showing an assert at startup.

Something to do with floppy drives: flopdrv.c line 123
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 07:45 AM

SVN r6381: N64 driver

I ported over all of angrylion et al's fixes from their MESS-derived plugin. Things look considerably better blending-wise, but unfortunately, texture seam issues remain. frown

Some noteworthy improvements:

MAME now runs in MESS via the N64 driver:







Mario 64's trees, shadows, and decal textures look much, much better thanks to the perfect Z handling:





Likewise, Mario's textures are finally correct thanks to the updated Color Combiner and Blender handling:


Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 07:54 AM

SVN r6381, continued:
Dr. Mario 64 actually displays the little swaying guy:



Paper Mario's file dialogues look much better:



Paper Mario also kindly deigns to provide us with a concrete, verbose example where the MIPS core's TLB handling craps the bed, that we can point Aaron at:



The balloons, alpha cutout, and Z buffering are finally correct in Mario Kart 64:



Tetrisphere's block colors now show up, rather than simply being a single diffuse color:



Also, can you please inform me if you see a compile failure with this SVN revision? I am not seeing anything of the sort with what I thought were the latest MAME / MESS build tools, so I'm at a loss.

Also, if anyone who has a 64-bit machine and has the skills to repair the N64 driver on 64 bits, feel free...
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 08:55 AM

Everyone please forgive the temporarily broken build, depending on the build settings you use. I'm working on a fix as fast as possible...

ETA: More WIP shots, as I make them:

Zelda 64, the weird-looking moon has been fixed:



r6386:

Zelda 64, title appears correct:



Beetle Adventure Racing, EA logo appears correct:



Super Smash Bros., title screen appears correct:

Posted By: angrylion

Re: SVN builds - new driver flood - 11/15/09 10:26 AM

Concerning r6385, things that are apparent from the svn log:
hresdiff should equal hres now, no need to shift;
pixdiff should always be 1 now;
replace ^1 by ^WORD_ADDR_XOR or something;
it's better to eliminate Lsw, hresdiff and pixdiff;
no more need for trickery like

leftuppix = ((leftuppix ^ 2) + 2) ^ 2;

in case you still have it, just

leftuppix++;
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 10:39 AM

Originally Posted By angrylion
Concerning r6385, things that are apparent from the svn log:
hresdiff should equal hres now, no need to shift;
pixdiff should always be 1 now;
replace ^1 by ^WORD_ADDR_XOR or something;
it's better to eliminate Lsw, hresdiff and pixdiff;
no more need for trickery like

leftuppix = ((leftuppix ^ 2) + 2) ^ 2;

in case you still have it, just

leftuppix++;


Cool, I'll submit an update to that effect shortly, please let me know if I've done it correctly.

Edit: Done, let me know if it looks right to you.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 11:11 AM

SVN r6389: N64 driver

And now, an unusual type of WIP. Let's call it:

N64 Driver: What It Can't Do

It can't display Bio F.R.E.A.K.S.'s text properly.



It can't display the background images in Mario Kart 64 properly.



It messes up on some of the UI art in Automobili Lamborghini.



It has a strangely offset screen in The New Tetris and the Rampage games.



It hangs on the first frame in Banjo-Kazooie.



It also can't boot a number of games; those including Chameleon Twist, Perfect Dark, GoldenEye 007, Banjo-Tooie, Conker's Bad Fur Day, Resident Evil 2, Indiana Jones and the Infernal Machine, Star Wars Episode 1 - Battle for Naboo, Yoshi's Story, and many others.

At this point, I think it might be best for anyone interested in contributing to the N64 driver to look into the main CPU or machine/n64.c-related issues keeping these games from booting. smile
Posted By: angrylion

Re: SVN builds - new driver flood - 11/15/09 11:24 AM

Concerning r6388, I think video_filter16 should look like this:
INT32 leftuppix = - hres - 2;//not -4 now, it's in pixels now, not bytes
//and so on
...
INT32 leftupcvg = - hres - 2;//still -2
...
pix = vbuff[leftuppix ^ WORD_ADDR_XOR];
cvg = hbuff[leftupcvg ^ BYTE_ADDR_XOR] & 3;
...
leftupcvg++;//these two may be merged now
leftuppix++;
Posted By: angrylion

Re: SVN builds - new driver flood - 11/15/09 11:39 AM

Concerning r6387, haven't we already determined that these registers are 24-bit wide?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 11:40 AM

Originally Posted By angrylion
Concerning r6387, haven't we already determined that these registers are 24-bit wide?


I don't know, why don't you hop onto IRC and talk to DrgnMndd about it? Not to disbelieve docs or anything like that, but r6387 was suggested by him based on behavior observed with real hardware.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 11:41 AM

Eta: Double post, please delete
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/15/09 01:04 PM

Originally Posted By Anna Wu
SVN r6350
Victor/Interact/Hector driver

Victor



Interact



Hector1



Hec2hrp



Hec2mx80



PS: I ask Daniel Coulom for a tool to convert .k7 to .wav format (tape). For the opposite way exist a tool.




Thanks goes to Yo_fr (the developer of the MESS driver) to send me a .wav files for testing.
He made a tool to convert K7 to WAV format, maybe he can help to add in castool.

hector1 driver
Game : Formule 1 (Tape)





hec2hr/hec2hrp/hec2hrx/hec2mx40/hec2mx80 driver
Game : Stramble (High resolution, Tape)


Posted By: Justin

Re: SVN builds - new driver flood - 11/15/09 07:14 PM

On a 64-bit MSVC compile, most N64 games (e.g. Mario 64) crash immediately.

The crash appears to be in video_filter16, at the following line:
pix = vbuff[leftup ^ WORD_ADDR_XOR];

leftup is 0xfffffebe. WORD_ADDR_XOR is a macro so I can't easily debug what it's returning but maybe it's wrong on 64-bit?

EDIT: Hmm no, seems to be the same as on 32-bit, I guess leftup or vbuff is wrong somehow....
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 07:33 PM

Originally Posted By Justin
leftup is 0xfffffebe. WORD_ADDR_XOR is a macro so I can't easily debug what it's returning but maybe it's wrong on 64-bit?


No, I'm pretty sure that leftup is plenty bad enough without having to look at WORD_ADDR_XOR as a culprit. Unfortunately, I don't have a 64-bit machine so that I can take a look myself.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/15/09 07:37 PM

leftup being negative is all manner of wrong without the xor, yes. Probably because almost no variables in the new N64 RDP code are initialized ever, sayeth GCC 4.
Posted By: Justin

Re: SVN builds - new driver flood - 11/15/09 07:42 PM

Well, given that hres is 320, and leftup is set to -hres - 2, I'm not sure what other result you can expect....
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/15/09 08:09 PM

Me neither, but using a negative number as an array index is always wrong.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/15/09 08:12 PM

Oh, god. This code is beyond Acho levels of 64-bit dirtiness. If MSVC compiled it without warnings the compiler team should be nuked from orbit.

Code:
Compiling src/mame/video/n64.c...                                 
cc1: warnings being treated as errors                             
src/mame/video/n64.c: In function ‘video_update_n64’:             
src/mame/video/n64.c:565: error: cast from pointer to integer of different size
src/mame/video/n64.c:565: error: cast from pointer to integer of different size
src/mame/video/n64.c: In function ‘video_filter16’:
src/mame/video/n64.c:2139: error: cast from pointer to integer of different size
src/mame/video/n64.c:2165: error: cast from pointer to integer of different size
src/mame/video/n64.c:2172: error: cast to pointer from integer of different size
src/mame/video/n64.c:2173: error: cast to pointer from integer of different size
src/mame/video/n64.c:2199: error: cast to pointer from integer of different size
src/mame/video/n64.c:2200: error: cast to pointer from integer of different size
src/mame/video/n64.c:2226: error: cast to pointer from integer of different size
src/mame/video/n64.c:2227: error: cast to pointer from integer of different size
src/mame/video/n64.c: In function ‘divot_filter16’:
src/mame/video/n64.c:2356: error: cast from pointer to integer of different size
src/mame/video/n64.c:2358: error: cast to pointer from integer of different size
src/mame/video/n64.c:2358: error: cast to pointer from integer of different size
src/mame/video/n64.c:2359: error: cast to pointer from integer of different size
src/mame/video/n64.c:2359: error: cast to pointer from integer of different size
src/mame/video/n64.c: In function ‘restore_filter16’:
src/mame/video/n64.c:2449: error: cast from pointer to integer of different size
src/mame/video/n64.c:2468: error: cast to pointer from integer of different size
src/mame/video/n64.c:2504: error: cast to pointer from integer of different size
src/mame/video/n64.c:2541: error: cast to pointer from integer of different size
Posted By: Justin

Re: SVN builds - new driver flood - 11/15/09 08:21 PM

Those cast issues should have been fixed in MESS SVN last night. I do still get some different warnings in MSVC though:

Quote:
src\mame\video\n64.c(3488) : warning C4819: The file contains a character that cannot be represented in the current code page (932). Save the file in Unicode format to prevent data loss
c:\games\emulated\mess\src\mame\video\n64.c(4271) : warning C4701: potentially uninitialized local variable 'dsdeh' used
c:\games\emulated\mess\src\mame\video\n64.c(4272) : warning C4701: potentially uninitialized local variable 'dtdeh' used
c:\games\emulated\mess\src\mame\video\n64.c(4273) : warning C4701: potentially uninitialized local variable 'dwdeh' used
c:\games\emulated\mess\src\mame\video\n64.c(4274) : warning C4701: potentially uninitialized local variable 'drdeh' used
c:\games\emulated\mess\src\mame\video\n64.c(4275) : warning C4701: potentially uninitialized local variable 'dgdeh' used
c:\games\emulated\mess\src\mame\video\n64.c(4276) : warning C4701: potentially uninitialized local variable 'dbdeh' used
c:\games\emulated\mess\src\mame\video\n64.c(4277) : warning C4701: potentially uninitialized local variable 'dadeh' used
c:\games\emulated\mess\src\mame\video\n64.c(4278) : warning C4701: potentially uninitialized local variable 'dzdeh' used
c:\games\emulated\mess\src\mame\video\n64.c(1859) : warning C4701: potentially uninitialized local variable 't0' used
c:\games\emulated\mess\src\mame\video\n64.c(1864) : warning C4701: potentially uninitialized local variable 't3' used


(The Unicode thing is because of a comment in what appears to be the legacy Windows Russian encoding on line 4008. I'd convert it myself if it wasn't in a src/mame/ file.)
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/15/09 08:50 PM

Oh, well those cast issues weren't fixed in MAME - that's current MAME SVN I got those errors from. Could whoever fixed them in MESS please submit them over?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/15/09 09:05 PM

Originally Posted By R. Belmont
Oh, well those cast issues weren't fixed in MAME - that's current MAME SVN I got those errors from. Could whoever fixed them in MESS please submit them over?


I will as soon as I fix all of these Mysterious Issues That I'm Not Seeing in the MESS build, then I'll move it all over en masse.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/16/09 04:51 AM

Incidentally, current SVN does compile finally, but zero (0) N64 games don't crash on a 64-bit build now, which makes it impossible for me or Aaron to diagnose your TLB problems. Can we back out to 0.135 until angrylion and friends get their shit together?
Posted By: angrylion

Re: SVN builds - new driver flood - 11/16/09 10:54 AM

Yeah, my code is dirty in many aspects, not warning-free and not 64-bit compatible. I warned JD about that.
Negative array indices are allowed by C (C 99 standard, 6.5.2.1, 6.5.6.8), provided that an array is effectively a pointer into another array, which is the case here. It's a dangerous programming practice, though.
leftup and the likes should be signed, as I suggested on the previous page. In r6390 JD made them unsigned, it may be (just a guess) a cause for crashes on 64 bit machines.
One option is to exclude VI filters entirely until the clean implementation is written.
Another option is to roll back altogether and then add functionality from my code in parts.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/16/09 01:27 PM

I just checked in a potential fix for the 64-bit crashiness, see if that fixes anything.
Posted By: Justin

Re: SVN builds - new driver flood - 11/16/09 03:17 PM

Yes, that did the trick.
Posted By: Haze

Re: SVN builds - new driver flood - 11/16/09 07:10 PM

Originally Posted By angrylion
I warned JD about that.
Negative array indices are allowed by C (C 99 standard, 6.5.2.1, 6.5.6.8), provided that an array is effectively a pointer into another array, which is the case here. It's a dangerous programming practice, though.


I'm pretty sure we used to have a bug in a few drivers caused by this very thing on 64-bit systems, and when compiled as 64-bit a negative array index would kill it on the spot, regardless of how valid it was. Might have been an issue with that version of the compiler tho.

Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/16/09 07:48 PM

Fortunately, that doesn't seem to happen anymore. INT32-ing the UINT32s took care of it nicely. Thanks for the suggestion, AL! smile
Posted By: Lord Nightmare

Re: SVN builds - new driver flood - 11/18/09 12:07 PM

Gamegear audio is now stereo (finally!). Thanks to Justin Kerk for prodding me to get it done, as well as providing recordings from his gamegear which I used to tweak the PRNG with (it still may be offset in time by one bit...) Please let me know if anyone runs into any issues with this or any of the other chips covered by sn76496.c

LN
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/18/09 11:58 PM

Originally Posted By R. Belmont
r6197 adds a -createuncomphd option to chdman. It works identically to -createhd but the resulting CHD is not compressed and not write protected. Would everyone who complained about this (mizapf, I'm looking at you) please test this before I throw it over to MAME?


Any chance of a -createblankuncomphd, for those of us who are trying to bring up a new machine from scratch? Also, what is this mystical "inputoffs" thing? It isn't needed by -createblankhd, so what is it?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/22/09 09:26 AM

SVN r6437
compilation problem

H:\SVN\Bin\mess>mingw32-make TARGET=mess
Compiling src/mess/drivers/mac.c...
cc1.exe: warnings being treated as errors
src/mess/drivers/mac.c:614: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:615: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:616: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:617: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:618: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:619: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:620: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:621: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:622: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:624: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:625: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:626: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:627: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:628: warning: integer constant is too large for 'long' type
src/mess/drivers/mac.c:631: warning: integer constant is too large for 'long' type

mingw32-make: *** [obj/windows/mess/mess/drivers/mac.o] Error 1
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/22/09 12:20 PM

r6435: SGI Indy (IP22) driver

IRIX now starts to boot in order to launch the installer. Click the images for bigger versions.




Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/22/09 12:25 PM

JD, how big (uncompressed)is the IRIX Installation CD ?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 11/22/09 02:32 PM

The full IRIX set is like 7 CDs I think.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/22/09 03:10 PM

Originally Posted By R. Belmont
The full IRIX set is like 7 CDs I think.


Thank you for the info, Arbee. smile

Can you please check the reason ?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/22/09 05:43 PM

Originally Posted By Anna Wu
Originally Posted By R. Belmont
The full IRIX set is like 7 CDs I think.


Thank you for the info, Arbee. smile


That said, you only need the first disc to actually launch the installer or launch diagnostics, both of which currently fail MESS, so downloading all 7 CDs at this point would be... overly optimistic. wink
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/22/09 05:52 PM

Originally Posted By Anna Wu
Originally Posted By R. Belmont
The full IRIX set is like 7 CDs I think.


Thank you for the info, Arbee. smile

Can you please check the reason ?


SVN r6438
Thanks for the fix (32-bit compiles) smile
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 11/23/09 11:22 PM

I found and fixed a bug in lib/util/aviio.c. Without this fix, aviio.c can write out ridiculous numbers for the "total length" field of audio streams. It could be that MAME and/or MESS never trigger this bug, but CD-i Emulator certainly did.

You can find my one-line fix here: http://www.cdiemu.org/download/aviio.diff

Can somebody with SVN access please apply this? I've also sent it to Aaron for inclusion in MAME.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/23/09 11:39 PM

Quick question: Should the condition in the if just before it:

Code:
 		/* add up the samples */
 		if (channelsamples > 0)


be changed to:

Code:
 		/* add up the samples */
 		if (chunksamples > 0)


Since channelsamples was changed to chunksamples in your diff on the line beneath it?
Posted By: CD-i Fan

Re: SVN builds - new driver flood - 11/23/09 11:41 PM

Very good question, and I think the answer is yes, which means that the "if" could just as well be deleted because chunksamples is always > 0.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/23/09 11:49 PM

Originally Posted By CD-i Fan
Very good question, and I think the answer is yes, which means that the "if" could just as well be deleted because chunksamples is always > 0.


In that case, it looks good to me.

In other news, SVN r6449 should have all of the changes necessary for the SGI Indy driver (ip225015) to actually boot the Integrated Diagnostic Environment off of a CD (and promptly fail the diagnostics), or to boot sash (the SGI Stand-Alone Shell) and fx (the disc-partitioning tool), partition a blank HDD image, write out a valid boot sector, then boot up the IRIX 6.5 installer, at which point the installer bombs out. But still, progress.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/24/09 05:10 AM

SVN r6447
" The Thompson artwork files are now zipped up. "

The artwork file called now " Thompson.zip " instead of " Thomson.zip ", is this right ?

Posted By: Robbbert

Re: SVN builds - new driver flood - 11/24/09 09:02 AM

Oops, wrong filename. Fixed now. grin
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/24/09 09:52 AM

Originally Posted By robbbert
Oops, wrong filename. Fixed now. grin


Thanks, robbbert. smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 11/27/09 12:14 PM

SVN r6472 adds sound to the Pocket Station driver. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 11/27/09 02:50 PM

SVN (own build )

Sord M5 driver
Game : Zac Banic (tape) and Solitaire (tape) + Basic-I (cartridge).

Zac Banic, loaded with command " Tape "



Solitaire, loaded with command " Chain "

Posted By: CD-i Fan

Re: SVN builds - new driver flood - 11/29/09 09:35 AM

Originally Posted By Just Desserts
Originally Posted By CD-i Fan
Very good question, and I think the answer is yes, which means that the "if" could just as well be deleted because chunksamples is always > 0.

In that case, it looks good to me.

I'd like to modify my previous answer. If you want correctly terminated audio streams it needs to be
Code:
		/* add up the samples */
		if (channelsamples > chunksamples)
			file->info.audio_numsamples = stream->samples += chunksamples;
		else if (channelsamples > 0)
			file->info.audio_numsamples = stream->samples += channelsamples;

The previous code will count extra silence at the end of the audio stream, the amount depending on the wether you left in the original if or modified it / took it out.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 12/01/09 06:40 AM

SVN r6489: SGI IRIS 3130 IP2 driver

Code:
C:\mingw2\mess_latest>messd sgi_ip2 -r 640x480 -window -nonewui -nodebug -nolog
-nosound
&#9658;IRIS (IP2 - Revision E) Monitor Version 3.0.10 July 1, 1987
Unknown boot type 0x5
Power has been lost to the machine
iris> help
.:help: Error 0
iris> printenv
.:printenv: Error 0
iris> hinv
.:hinv: Error 0
iris> wank
.:wank: Error 0
iris> arse
.:arse: Error 0
iris> quit
.:quit: Error 0
iris> exit
PROMs exiting
&#9658;IRIS (IP2 - Revision E) Monitor Version 3.0.10 July 1, 1987
Unknown boot type 0x5
Power has been lost to the machine
iris> Average speed: 47.02% (14 seconds)


Wish I knew what "Error 0" meant, but it's a start. smile
Posted By: judge

Re: SVN builds - new driver flood - 12/01/09 06:49 AM

it probably means it has no power :P
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 12/01/09 11:47 AM

No, it's just a generic Open Firmware error when it can't run a command. The "lost power" thing means the NVRAM checksum failed on boot or it believes for some other reason that the machine wasn't shut down gracefully.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 12/01/09 01:48 PM

SVN r6492: SGI IRIS 3130 IP2 driver

After upping the amount of visible memory and adding 64k for a "BSS" segment for the PROM (whatever that means), it gets a bit chattier and a bit more polite:

Code:
&#9658;IRIS (IP2 - Revision E) Monitor Version 3.0.10 July 1, 1987
Memory Size 1mb (Physical Map (1mb/bit) 0x00000001)
Configuration Switch: 0x9805
  Multibus Window (2mb) at Megabytes 2 and 3.
  Multibus accessible memory (1mb) begins
    at Physical memory page 0,
    at Virtual address 2000000.
  WARNING: Multibus accessible memory overlaps all
           Physical memory
Power has been lost to the machine
iris> h
General Monitor Commands (All numeric values in hex):
  MEDIA is one of the following:
      hd        - hard disk.(look for ip, sd, then md)
      ct        - cartridge tape.(look for st, then mt)
      fd        - floppy disk (look for sf, then mf)
      ip        - interphase disk.
      sd(or si) - storager disk.
      md        - midas disk.
      st(or sq) - storager cartridge tape.
      mt(or mq) - midas cartridge tape.
      sf        - storager floppy disk.
      mf        - midas floppy disk.
      tcp       - TCP network.
      xns       - XNS network.
      rom       - EPROM board.
  DEVSPEC is the one of the following:
      host name         - name of a host. (MEDIA must be tcp or xns.)
      unit              - unit number of device (0, 1, ...).
                            (MEDIA must be a tape or disk device.)
      <unit><fs>        - unit number and filesystem (a - h).
                            (MEDIA must be a disk device.)
      address           - multibus address.
                            (MEDIA must be the EPROM board.)

Continue (y or n)?: y
[MEDIA.DEVSPEC:][file]  load and begin execution of the named file.
                                    file defaults to defaultboot
                                    SPECs are from switch settings

b [MEDIA.DEVSPEC:][file]        same as above.

t [DEVSPEC:][file]              same as b with MEDIA = tcp.

n [DEVSPEC:][file]              same as b with MEDIA = xns.

ls [MEDIA.DEVSPEC:][file]       list the files on the device.

l [MEDIA.DEVSPEC:][file]        load but don't begin execution of the file.

g address [stack]               start executing at specified address.
                                the stack address is optional.

h|?                             print this help message.

Continue (y or n)?: y
set                             print the current set values.
set media MEDIA                 set the default boot media.
set devspec DEVSPEC             set the default boot device spec.
set debug 0/1                   set the debug mode.
set display                     set display options from switch settings.
set dcr BITS OPTION             set DC4 bits and option.
set inetaddr ADDRESS            set TCP/IP address of this machine.
set servaddr ADDRESS            set TCP/IP address of boot server.
set gateaddr ADDRESS            set TCP/IP address of boot gateway.
                                32-bit TCP/IP ADDRESS is given as four bytes
                                in decimal separated by dots (89.0.0.1)

exit                            reset the PROMS.

fm{b|w|l} ADDR VALUE [INCR] [CNT]
                                fill memory as byte, word or long
                                starting at ADDR, with initial VALUE,
                                incrementing VALUE by INCR for CNT times.
                                (INCR defaults to 0; CNT defaults to 1)

dm{b|w|l} ADDR [CNT]            display memory as byte, word or long.
                                (CNT defaults to 1)

em{b|w|l} ADDR                  edit memory interactively as byte,
                                word or long.

dpr                             display processor registers.

epr REGISTER                    edit the given processor register.
                                (sr, vbr, cacr, caar, sfc, dfc).

iris> set
Clock is running:
     Time      Alarms       Date        Registers
   04:02:00   05:03:01   08/07[6]/09   0a 0b 0c 0d
Saved RAM: magic 0xe0f1011, boot 0x12, power 0x13, dcr 0x15
Boot Environment:
   Media  : mon
   Devspec:
TCP/IP Network Boot Environment:
   Internet address of this machine: 28.29.30.31 (from Saved RAM)
   Internet address of boot server:  (not set)
   Internet address of boot gateway: (not set)
Misc Environment:
   Version: 3.0.10 July 1, 1987
   Serial Number: 6426
   IP2 Revision E with 1 Meg of memory
   FPA is installed
   Ethernet controller is not installed
iris>

Posted By: Multipass

Re: SVN builds - new driver flood - 12/05/09 08:32 PM

Hello all,
I have an error with last svn compil:

Compiling src/mame/machine/konamigx.c...
src/mame/machine/konamigx.c: In function 'konamigx_esc_alert':
src/mame/machine/konamigx.c:271: error: 'K053247_ram' undeclared (first use in this function)
src/mame/machine/konamigx.c:271: error: (Each undeclared identifier is reported only once
src/mame/machine/konamigx.c:271: error: for each function it appears in.)
mingw32-make: *** [obj/windows/mame/mame/machine/konamigx.o] Error 1

Please why i have this, and someone can fix this?
More thanks bye

Edit: Fixed! sorry for inconvenient wink
Posted By: mahlemiut

Re: SVN builds - new driver flood - 12/07/09 02:05 AM

Hmmm.... wink



Using Ishido, which is basically runs a minimal TownsOS setup and the game is run within that. Still a ways to go, but it's progress. smile

TownsOS is basically a protected mode GUI on top of MS-DOS 3.1. (Sounds like Windows 3.x :))
Posted By: R. Belmont

Re: SVN builds - new driver flood - 12/07/09 03:04 AM

Ooh, nice smile
Posted By: Kale

Re: SVN builds - new driver flood - 12/07/09 03:19 AM

Is Ishido that mildly interesting puzzle game that's also on GameBoy, PC and Megadrive? shocked
Posted By: mahlemiut

Re: SVN builds - new driver flood - 12/07/09 03:55 AM

Originally Posted By Kale
Is Ishido that mildly interesting puzzle game that's also on GameBoy, PC and Megadrive? shocked

Probably smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/07/09 11:50 AM

SVN r6566
TinyGiant HT68K driver
K-OS (Floppy Disk)

Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/12/09 06:16 PM

Compile (32-bit) problem

SVN r6608 +

Compiling src/mess/machine/c1541.c...
cc1.exe: warnings being treated as errors
src/mess/machine/c1541.c: In function 'bit_tick':
src/mess/machine/c1541.c:240: warning: 'byte_ready' is used uninitialized in this function
mingw32-make: *** [obj/windows/mess/mess/machine/c1541.o] Error 1

Posted By: R. Belmont

Re: SVN builds - new driver flood - 12/12/09 07:01 PM

Yeah, I posted about that earlier. It's a bit of a nonsense statement so I'm not sure what Curt intended.
Posted By: Curt Coder

Re: SVN builds - new driver flood - 12/12/09 08:42 PM

Oops, it's a copypaste error. Fixed to svn.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/12/09 11:56 PM

Thanks for the fix. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/13/09 10:22 PM

Compile (32-bit) problem

SVN r6620 +

Compiling src/mess/machine/c1541.c...
cc1.exe: warnings being treated as errors
src/mess/machine/c1541.c: In function 'cbmserial_reset_c1541':
src/mess/machine/c1541.c:278: warning: unused variable 'c1541'
mingw32-make: *** [obj/windows/mess/mess/machine/c1541.o] Error 1

Posted By: mahlemiut

Re: SVN builds - new driver flood - 12/17/09 09:21 PM

More progress. wink



Not sure why it doesn't clear the foreground layer at this point...
Posted By: R. Belmont

Re: SVN builds - new driver flood - 12/17/09 09:26 PM

Sweet smile
Posted By: Kale

Re: SVN builds - new driver flood - 12/18/09 01:30 AM

Originally Posted By mahlemiut

Not sure why it doesn't clear the foreground layer at this point...


Maybe a missing clear VRAM register or something similar?
Posted By: mahlemiut

Re: SVN builds - new driver flood - 12/18/09 03:02 AM

It should not clear the VRAM all at once, so that doesn't seem likely.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/18/09 06:51 AM

Very nice, Barry smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 12/20/09 02:10 AM

RDP optimizations, enjoy 'em now. smile

Before:
Code:
C:\mingw\mess>mess n64 -cart ".\software\Super Mario 64 (U) [!].v64" -str 30
Average speed: 22.81% (29 seconds)

C:\mingw\mess>mess n64 -cart ".\software\Bust-A-Move 2 - Arcade Edition (U) [!].zip" -str 30
Average speed: 112.42% (29 seconds)

C:\mingw\mess>mess n64 -cart ".\software\Mario Kart 64 (U) [!].zip" -str 30
Average speed: 28.74% (29 seconds)


After:
Code:
C:\mingw\mess>mess n64 -cart ".\software\Super Mario 64 (U) [!].v64" -str 30
Average speed: 25.98% (29 seconds)

C:\mingw\mess>mess n64 -cart ".\software\Bust-A-Move 2 - Arcade Edition (U) [!].zip" -str 30
Average speed: 119.68% (29 seconds)

C:\mingw\mess>mess n64 -cart ".\software\Mario Kart 64 (U) [!].zip" -str 30
Average speed: 33.57% (29 seconds)

Posted By: Multipass

Re: SVN builds - new driver flood - 12/20/09 11:10 PM

Hello all i have some errors with last svn 6716:

src/mess/drivers/n64.c:41: error: initializer element is not constant
src/mess/drivers/n64.c:41: error: (near initialization for 'address_map_n64_map[32].i')
src/mess/drivers/n64.c:42: error: initializer element is not constant
src/mess/drivers/n64.c:42: error: (near initialization for 'address_map_n64_map[44].i')
src/mess/drivers/n64.c:58: error: initializer element is not constant
src/mess/drivers/n64.c:58: error: (near initialization for 'address_map_rsp_map[10].i')
src/mess/drivers/n64.c:59: error: initializer element is not constant
src/mess/drivers/n64.c:59: error: (near initialization for 'address_map_rsp_map[20].i')
src/mess/drivers/n64.c:60: error: initializer element is not constant
src/mess/drivers/n64.c:60: error: (near initialization for 'address_map_rsp_map[30].i')
src/mess/drivers/n64.c:61: error: initializer element is not constant
src/mess/drivers/n64.c:61: error: (near initialization for 'address_map_rsp_map[40].i')
mingw32-make: *** [obj/windows/mamep/mess/drivers/n64.o] Error 1

I have tested clean compil but nothin
Someone know where is the problem please?
Thanks
Posted By: R. Belmont

Re: SVN builds - new driver flood - 12/21/09 01:08 AM

SVN #6716 works for me, and the automated builder passed as well. You're either doing something wrong or building on an exotic configuration - good luck figuring it out smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/25/09 10:53 AM

SVN r6736
Micronique Hector HRX driver
Game : Pengo (Cassette, .k7)



Thank you for your nice email, Jean-Jacques. smile
Posted By: mahlemiut

Re: SVN builds - new driver flood - 12/25/09 12:39 PM

Happy Boxing Day! (Well, it is past 1:30am here... smile





It's somewhat playable (joystick directions aren't quite right), has no sound effects (but there is CD-DA music), and still some sprite mis-alignment. But eh, it runs. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/25/09 04:42 PM

Originally Posted By mahlemiut
Happy Boxing Day! (Well, it is past 1:30am here... smile

It's somewhat playable (joystick directions aren't quite right), has no sound effects (but there is CD-DA music), and still some sprite mis-alignment. But eh, it runs. smile


Thank you for the info, Barry ! smile
I will test the driver as soon as possible. At moment is my clean compile time.
Except Raiden, any other game (CD or Floppy Disk) can be loaded ?
Posted By: mahlemiut

Re: SVN builds - new driver flood - 12/25/09 08:44 PM

Originally Posted By Anna Wu
I will test the driver as soon as possible. At moment is my clean compile time.
Except Raiden, any other game (CD or Floppy Disk) can be loaded ?

Nope, pretty much limited to real-mode DOS applications. Most FM Towns software requires protected mode. I can't even run simple protected mode test programs that I've written using Towns-GCC (GCC 2.45 cross-compiler, generates protected mode .EXP executables)
Maybe there are other real-mode games out there, but I'm sure most require protected mode.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 12/26/09 04:02 AM

Can we get a discussion going on the desirability of the "F4 support" patches that Robbbbbert is submitting? I vote "wastes memory and execution time and clutters the drivers for no useful reason", but MESS has never had adult supervision before so I assume they will continue to go in.

In case it's not clear: F4 is a developer aid, not a user feature. It not working for a driver is not considered a problem. Since the drivers the support is being added to all already can decode and display the characters properly there is no developer benefit to them either.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 12/26/09 04:19 PM

Me, I added a few debugging dips to the CD-i driver so that it would be easier for people to rip graphics from the game, so I don't see a problem with what Robbbert's doing.

While it can be argued that it "clutters up" a driver, you said yourself that the drivers can already decode and display the characters properly, so that implies that the drivers are already largely complete. In the event that they are not, throwing in a few more blocks of code isn't going to harm anything.

While you claim that it's "a developer aid, not a user feature," users already view F4 as a user feature, and you as a MAME dev should know quite well how much official proclamations from devs have an impact on what users think. At this point, I'd say we'd might as well run with it.

About the only suggestion I have, not having seen the code that Robbbert is adding in (the above "few more blocks of code" was an assumption), is that it should be commented as being related to the F4 viewer and not the actual functional emulation of the system.
Posted By: Haze

Re: SVN builds - new driver flood - 12/26/09 05:16 PM

Originally Posted By R. Belmont
Can we get a discussion going on the desirability of the "F4 support" patches that Robbbbbert is submitting? I vote "wastes memory and execution time and clutters the drivers for no useful reason", but MESS has never had adult supervision before so I assume they will continue to go in.

In case it's not clear: F4 is a developer aid, not a user feature. It not working for a driver is not considered a problem. Since the drivers the support is being added to all already can decode and display the characters properly there is no developer benefit to them either.


Context.

While I agree, MESS needs more overall quality control you have to look at things in context.

I looked over some of the changes, and they have no performance hit. This is case #1, they're a one time decode, which decode the character stored in the BIOS rom.

Case #2 is dynamic decodes (which decode characters as they're uploaded to RAM) *do* have a cost, especially if they're not needed by the driver, which obviously if they're added as an afterthought aren't.

Of course in case #1 you'll only ever get the same decode every time you run the driver. The characters never change, they're the bios characters, they're not very interesting*

In case #2, the characters will change. If you're developing a driver, this might be useful to indicate how far a specific game has gone towards loading, has it ever uploaded gfx data, or did it get stuck before that point? It can be considered a useful dev aid, if the developers of that driver need it.

As with MAME, there are 2 groups of users. Developers using the code, to make improvements, and end-users, just using it to play games. Benefits to the first group can be very important, benefits to the 2nd group, a little less-so. You could argue that having F4 available (on dynamic drivers) benefits the developers (seeing what's uploaded) but puts players at a disdavantage (slower speed).

For changes like this that's what is important, you need to weigh up the cost / benefit, in some cases they're useless, in other cases not so much. The ones checked in so far are harmless, but ultimately not very interesting. They add no value, but they have no cost aside a few lines of easily understandable code.


* For console / computer systems where they'll likely only contain the BIOS font, obviously MAME is different where most of the game GFX are usually in static ROM.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 12/28/09 06:38 AM

With the latest SVN tree, there should be a number of considerable N64 optimizations, including a muchly-improved RSP DRC and more RDP flattening. The following caveats should be noted:

1) There is an as-yet unknown bug affecting texel lookups for certain texture types, as exhibited in Tetrisphere.
2) I have temporarily disabled vi_filter16 and divot_filter16, because there are a few fiddly things about them I don't trust. This has given a temporary speed boost, but expect a slight drop in performance once they come back online.
Posted By: AaronGiles

Re: SVN builds - new driver flood - 12/28/09 07:02 AM

Don't forget to put your latest changes in MAME....
Posted By: Just Desserts

Re: SVN builds - new driver flood - 12/28/09 09:33 AM

I am, I was just making sure the MESS autobuilder was appeased before moving the changes on to MAME. Don't want to break the build, after all.
Posted By: belegdol

Re: SVN builds - new driver flood - 12/28/09 05:38 PM

rev 6807 segfaults on startup for me:
Quote:
(gdb) run n64 -cart ~/data/mess/software/n64/Super\ Mario\ 64\ \(U\)\ \[\!\].zip
Starting program: /home/jsikorski/cvs/mess/mess n64 -cart ~/data/mess/software/n64/Super\ Mario\ 64\ \(U\)\ \[\!\].zip
[Thread debugging using libthread_db enabled]

Program received signal SIGSEGV, Segmentation fault.
cpu_get_local_time (device=0x1) at src/emu/cpuexec.c:920
920 cpu_class_data *classdata = get_class_data(device);
(gdb) bt
#0 0x0000000000895c67 in cpu_get_local_time (device=0x1)
#1 0x0000000000895d38 in cpuexec_override_local_time (
machine=<value optimized out>, default_time=...) at src/emu/cpuexec.c:944
#2 0x00000000008ee69a in get_current_time (machine=0x7f59400)
at src/emu/timer.c:144
#3 _timer_alloc_common (machine=0x7f59400) at src/emu/timer.c:625
#4 _timer_alloc_internal (machine=0x7f59400) at src/emu/timer.c:659
#5 0x00000000008c95f9 in init_machine (machine=<value optimized out>)
at src/emu/mame.c:1505
#6 mame_execute (machine=<value optimized out>) at src/emu/mame.c:324
#7 0x00000000008950b8 in cli_execute (argc=18462024,
argv=<value optimized out>, osd_options=0x7f21250)
at src/emu/clifront.c:170
#8 0x0000000000862f52 in main (argc=4, argv=0x7fffffffe0f8)
at src/osd/sdl/sdlmain.c:392

This seems to happen with other drivers as well (I tried genesis, gameboy and saturn). This is on Fedora 12 x86_64.
Posted By: Duke

Re: SVN builds - new driver flood - 12/28/09 07:33 PM

Did you do a clean build?
Posted By: belegdol

Re: SVN builds - new driver flood - 12/28/09 07:49 PM

I did, and it helped. Are there any guidelines when a clean build is needed? Or is it just "try it if it breaks"?
Posted By: Firewave

Re: SVN builds - new driver flood - 12/29/09 05:18 AM

Always do a clean build or get a pre-compiled binary from bobz or somewhere else.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 12/29/09 05:26 AM

As a rule of thumb, if anything under src/emu or the makefile changes do a clean build. That doesn't cover all cases, but it's a reasonable heuristic.
Posted By: bobz

Re: SVN builds - new driver flood - 12/29/09 09:20 AM

Be aware that the pre-compiled binaries on my site are not always clean built. They might be broken too.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 12/30/09 04:36 PM

SVN r6799
RM Nimbus Driver
(update to RM nimbus driver, it will now boot from floppy and the keyboard now works, though the video is still buggy.)

Found a disk image called WELC.FLP on Internet which content MS-DOS 3.10
I made following:
Formated a real floppy disk (720KB) on normal PC (Win98 because WinXP not support this format) and not on real RM Nimbus PC.
Used the tool " 2FLOPPY " to read WELC.FLP image and write on the formatted Floppy Disk.
Used WinImage to read the Floppy Disk (A:) and saved as WELC.IMA

Unfortunately, my created Floppy Disk image is not booting.
My questions are now, made I some mistakes, is the reason i need to format the 720KB Disk on RM Nimbus PC or is the reason the RM Nimbus driver ?

Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/01/10 11:05 AM

Originally Posted By mahlemiut
Originally Posted By Anna Wu
I will test the driver as soon as possible. At moment is my clean compile time.
Except Raiden, any other game (CD or Floppy Disk) can be loaded ?

Nope, pretty much limited to real-mode DOS applications. Most FM Towns software requires protected mode. I can't even run simple protected mode test programs that I've written using Towns-GCC (GCC 2.45 cross-compiler, generates protected mode .EXP executables)
Maybe there are other real-mode games out there, but I'm sure most require protected mode.


Found another game called " Prince of Persia (.chd)" which I can load. The sound is not bad but can not play the game because of some graphic problems.



Raiden looks good. smile

Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/01/10 11:21 AM

SVN r6833
RM Nimbus driver
Welcome disk with MS-DOS 3.10


Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/01/10 06:07 PM

SVN r6844
SCV skeleton driver

I know, it is not working yet but if i start the driver (ROM check passed) I get following error message:

src/mess/drivers/scv.c: scv must use DEF_STR(Pause)
src/mess/drivers/scv.c: scv_pal must use DEF_STR(Pause)

r6844 was compiled (GCC 32-bit) without errors.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/01/10 06:31 PM

Originally Posted By Anna Wu
SVN r6844
SCV skeleton driver

I know, it is not working yet but if i start the driver (ROM check passed) I get following error message:

src/mess/drivers/scv.c: scv must use DEF_STR(Pause)
src/mess/drivers/scv.c: scv_pal must use DEF_STR(Pause)

r6844 was compiled (GCC 32-bit) without errors.


Thank you for the fast fix (r6846), judge. smile
Posted By: mahlemiut

Re: SVN builds - new driver flood - 01/01/10 08:47 PM

Originally Posted By Anna Wu
Found another game called " Prince of Persia (.chd)" which I can load. The sound is not bad but can not play the game because of some graphic problems.


Neat. Will have a look for that.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/01/10 08:56 PM

Originally Posted By mahlemiut
Originally Posted By Anna Wu
Found another game called " Prince of Persia (.chd)" which I can load. The sound is not bad but can not play the game because of some graphic problems.


Neat. Will have a look for that.


With the sound I mean the music and not the sound effects. smile
Posted By: judge

Re: SVN builds - new driver flood - 01/01/10 09:24 PM

Wheeeee

Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/01/10 09:41 PM

Originally Posted By judge
Wheeeee



A good start ! smile
Posted By: plgDavid

Re: SVN builds - new driver flood - 01/02/10 08:11 PM

8|

wow. i guess i'll have to pursue my upD1771c analysis sooner.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/02/10 10:43 PM

SVN r6877
Super A`can/SCV driver

First impressions smile

Super A`Can


SCV


Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/03/10 11:21 AM

SVN r6896
Super A`can/SCV driver

Next impressions smile

Super A`Can


SCV
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/03/10 12:54 PM

Super A'Can in motion: http://www.youtube.com/watch?v=V-j0vYzxHsQ
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/03/10 12:59 PM

Super cool. Looks like some layer enables are missing, but it's certainly playable smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/04/10 07:58 PM

SVN r6914
AT586 driver
Bootable Harddisk

Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/04/10 09:04 PM

Everyone should check out the Super A'Can driver now... with Kale's latest fixes, it looks fantastic! About the only remaining issue is the ROZ layer and whatever crazy scaling they're doing with the A'Can logo. smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/04/10 09:21 PM

Originally Posted By Just Desserts
Everyone should check out the Super A'Can driver now... with Kale's latest fixes, it looks fantastic! About the only remaining issue is the ROZ layer and whatever crazy scaling they're doing with the A'Can logo. smile


Looks nice. smile
The sound is not supported yet, right ?

I hope, more cart dumps are coming soon ...
Posted By: Kale

Re: SVN builds - new driver flood - 01/05/10 02:06 AM

Super Dragon Force (Super A'Can)





HW looks pretty much versatile, Boom Zoo didn't use most of the features of this thing, tilemap paging, bpp select, tilemap banking... I think that I need all of the games to do educated guesses.

Also, Sonic Dragon doesn't work at all, really no idea about that at the current time...
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/05/10 03:30 AM

Would you like me to take a look at Sonic Dragon, or do you plan to look at it later? :P
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/05/10 03:52 AM

From the videos on YouTube I'd say Sonic Dragon not working is a feature smile
Posted By: judge

Re: SVN builds - new driver flood - 01/05/10 10:23 AM

Originally Posted By Kale
Also, Sonic Dragon doesn't work at all, really no idea about that at the current time...


There are a couple of things like:
Code:
move.l  #$26ca4, $ff7be4.l

in the code but they are never executed so it ends up with a bad address at $ff7be4 which causes it to break.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/05/10 07:47 PM

My last SVN versions + bobz last build
Super Game Boy (supergb) driver

If I start the emulation, I get a crash.
Can someone confirm this problem ?
Posted By: judge

Re: SVN builds - new driver flood - 01/05/10 07:51 PM

Confirmed & fixed.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/05/10 07:57 PM

Originally Posted By judge
Confirmed & fixed.


Checked & thanks smile
Posted By: Kale

Re: SVN builds - new driver flood - 01/06/10 05:40 PM

Played a little with debug tricks on the other game...identified that isn't actually Sonic Dragon but Super Taiwanese Baseball League





Sprites are missing/wrong because it uploads a G. Clinton-esque scatter-gather DMA list at e90030, weird because it does:
ff637c src f00010 dst 2f size b900 param

f00010 is the video DMA, and here's the title screen list:



I've told you it was funky, no?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/06/10 07:14 PM

Originally Posted By Kale
Sprites are missing/wrong because it uploads a G. Clinton-esque scatter-gather DMA list at e90030, weird because it does:
ff637c src f00010 dst 2f size b900 param

f00010 is the video DMA, and here's the title screen list:



I've told you it was funky, no?


Is it that DMA is wrong, or could the game be re-basing the sprite data area, as I described in my post from a few days ago?
Posted By: Kale

Re: SVN builds - new driver flood - 01/06/10 07:18 PM

As you can see from the title screen (and the garbage that it puts shortly after it), data is missing, so it's a weird way of doing scatter-gather
Posted By: Kale

Re: SVN builds - new driver flood - 01/06/10 09:08 PM

And...here's this pseudo scatter-gather implemented wink

Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/06/10 09:15 PM

Great process, Kale smile
Posted By: SteveMartin

Re: SVN builds - new driver flood - 01/09/10 02:05 AM

Wow, I didn't expect to suddenly see A'can emulation getting anywhere so soon. I wonder where these ROM dumps are coming from, maybe all games will be dumped soon? Save the extra rare Rebel Star. Doubt we'll be seeing that one anytime soon.
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/09/10 05:05 AM

SVN r6981, Vii driver (skeleton):

Posted By: Multipass

Re: SVN builds - new driver flood - 01/09/10 07:49 PM

Thanks JD for the news wink
Somes vii games are dumped actually please?
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/10/10 10:27 AM

SVN r6986: Vii skeleton driver: Base Vii software

http://www.youtube.com/watch?v=OaC-X6-RCsg
Posted By: Kale

Re: SVN builds - new driver flood - 01/10/10 05:54 PM

Originally Posted By SteveMartin
Wow, I didn't expect to suddenly see A'can emulation getting anywhere so soon. I wonder where these ROM dumps are coming from, maybe all games will be dumped soon? Save the extra rare Rebel Star. Doubt we'll be seeing that one anytime soon.


HW-wise is pretty easy to HLE (at least the basic stuff), it's really a rip-off of the SFC & MD characteristics glued together (windows, tilamap bases, mosaic effects etc.). It's not really a bad system per-se and it's certainly better than the two aforementioned consoles, you can declare 3 normal layers + 1 roz + sprites, and you can declare the bpp for each layer indipendently, the only bottleneck that I can see so far is that the paletteram really seems to have only 256 color entries, it probably have v/h beam position stuff like the MD but it's still somewhat limited, especially for a 1995 system.

I'm fairly sure that with decent programmers this thing could have given true videogames, at least in-par with the average fourth generation gamelist anyway. Currently instead they all looks rip-offs of something else, that Super Dragon Force for example looks a lot like Falcom's Ys imho.

The rest of the games will not be dumped soon, we still need to locate and dump anything minus 05, 07 and 11.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/11/10 04:13 AM

SVN 6998
Vii driver

New Impressions

Posted By: Stiletto

Re: SVN builds - new driver flood - 01/11/10 05:02 AM

Wall-E? Okay, those "Chintendo" guys are my kind of crazy. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/11/10 05:09 AM

Wall-E is actually not for the Vii, it's a TV Plug-n-Play on similar hardware, and still about 12 months from being kosher for MESS. It was a great help in debugging the CPU emulation though smile
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/11/10 05:19 AM

Originally Posted By R. Belmont
Wall-E is actually not for the Vii, it's a TV Plug-n-Play on similar hardware, and still about 12 months from being kosher for MESS. It was a great help in debugging the CPU emulation though smile


Yes smile

Posted By: Stiletto

Re: SVN builds - new driver flood - 01/11/10 05:47 AM

Hoo boy... shocked
Posted By: SteveMartin

Re: SVN builds - new driver flood - 01/11/10 06:46 AM

I just bought Sonic Dragon and C.U.G. Once I get them and have enough fun playing them on my system I'll be sending them off to Kevtris.

Look forward to anymore progress you make. I'm very enthusiastic about this, never thought I'd see the day smile
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 01/11/10 07:03 AM

I picked up two of those Plug 'n Play TV Games cheap at a garage sale--Mystery Machine and Spiderman 3. Does anyone know if that's the same hardware? I haven't even tried to look inside them.
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 01/11/10 07:07 AM

We'll probably want some sort of artwork support for those TV games when they get in--half the fun is the crazy looking controllers. Half the lack of fun is the uncomfortable controllers. smile
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/11/10 04:29 PM

The other half of the lack of fun is that the "arcade classics" are ports on dumbed-down hardware and the originals generally just suck. But yeah, artwork for the controllers is probably a must smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/11/10 04:47 PM

Originally Posted By ranger_lennier
I picked up two of those Plug 'n Play TV Games cheap at a garage sale--Mystery Machine and Spiderman 3. Does anyone know if that's the same hardware? I haven't even tried to look inside them.


Unfortunately, this one guy's résumé seems to be the only thing listing the hardware on which some of these games run, and he didn't work on either of those. That said:

Quote:

* Disney Fairies – GPAC800/Accelerometer
* Power Rangers – GPAC800/Accelerometer
* Spiderman Masked Menace – GPAC500
* Star Wars:Clone Wars - SPG243 Lightgun
* Leapfrog - SPG243 BaseROM
* Toy Story – SPG243
* Animal Art Studio - SPG243
* Nemo – SPG243
* Fantastic Four – SPG240
* Spiderman – SPG110/GPAC300


Other folks' résumés list:

Quote:

* Spongebob Squarepants Dilly Dabbler - SPG110
* Super Silly Makeover - SPG110
* World Poker Tour - SPG110

Posted By: Christina

Re: SVN builds - new driver flood - 01/11/10 05:24 PM

Hi!
Let me ask...
I saw one of that Vii consoles another day...
For me that runned under a rom into console, without cart...
Then...
Why this system needs a cart to runs?
Thanks for the answers and for a great work for the team!
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/11/10 06:49 PM

The built-in ROM is actually just an internal cartridge that gets switched out. I've already told JD it should be made into a BIOS that optional carts override, but he's lazy smile
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/11/10 06:50 PM

Originally Posted By Christina
Hi!
Let me ask...
I saw one of that Vii consoles another day...
For me that runned under a rom into console, without cart...
Then...
Why this system needs a cart to runs?
Thanks for the answers and for a great work for the team!


Well, as the driver is clearly marked as a skeleton driver and it's clearly marked as "not working", you shouldn't worry about any driver-related issues at this time.

Originally Posted By R. Belmont
The built-in ROM is actually just an internal cartridge that gets switched out. I've already told JD it should be made into a BIOS that optional carts override, but he's lazy smile


That too.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 01/12/10 02:04 AM

r7008
Fujitsu FM-Towns / FM-Towns Marty / Car Marty

Noticed that the memory-mapped ports at 0xcff80-0xcffff are mirrored in I/O ports 0xff80-0xffff.







Prince of Persia is now very playable, except that the Japanese text is wrong.

Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/12/10 05:23 AM

That's interesting - it looks like they reproduced the graphics from the Apple II Double Hi-Res screens complete with artifacting. Or is the graphics emulation not correct yet?
Posted By: mahlemiut

Re: SVN builds - new driver flood - 01/12/10 09:56 AM

Other than the wrong Japanese character being displayed in places, I think it's otherwise pretty much correct. While I'm not terribly familiar with the Apple II, this version reminds me of the X68000 version, just somewhat faster.
Posted By: mahlemiut

Re: SVN builds - new driver flood - 01/12/10 11:50 AM

r7016


Hopefully the text is now correct.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/12/10 04:17 PM

SVN r7012
batman driver
Game : Batman

Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/12/10 04:33 PM

Originally Posted By Anna Wu
SVN r7012
batman driver
Game : Batman



For anyone curious, it turned out once I got Batman booting that it's actually (c)2004, so it's more than eligible for inclusion in MESS. smile
Posted By: Kaylee

Re: SVN builds - new driver flood - 01/18/10 02:10 AM

svn 7054:

I get the following error:

Compiling src/mess/software/32x_cart.c...
src/mess/software/32x_cart.c:486:91: error: macro "SOFTWARE" Require 7 arguments only 6 is given
src/mess/software/32x_cart.c:486: error: 'SOFTWARE' Undeclared here (not in a function)
src/mess/software/32x_cart.c:489: error: expected '}; before '{' token
mingw32-make: ***[obj/windows/mess/mess/software/32x_cart.o]Error 1
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/18/10 02:23 AM

The auto-builder agrees.
Posted By: etabeta78

Re: SVN builds - new driver flood - 01/18/10 06:22 AM

d'oh... I had a comma switched with a quote...
also, I missed an 'n' in crvision_list

now it's building again. sorry.
Posted By: incog

Re: SVN builds - new driver flood - 01/21/10 05:36 PM

Judge recently added a Casio PV-1000 driver.



Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/21/10 05:37 PM

Very cool. I guess in-game doesn't work yet?
Posted By: incog

Re: SVN builds - new driver flood - 01/21/10 05:51 PM

Inputs are still not implimented, the only other emu for the PV-1000 (ePV-1000) also suffers from input bugs and you cant start the 8kb games (amidar, pooyan)
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/21/10 06:30 PM

Ahh, it's one of those things. Still, nice to see more Japanese computers running smile
Posted By: ranger_lennier

Re: SVN builds - new driver flood - 01/22/10 04:26 AM

I'm looking forward to seeing more of the PV-1000. Does the system have any sort of internal BIOS, or are the carts all you need?
Posted By: judge

Re: SVN builds - new driver flood - 01/22/10 07:16 AM

There is no BIOS; everything is on the carts.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/23/10 04:45 AM

With SVN 7137, the .D64 images for Second Reality C64 don't load anything (they didn't with the old HLE'd 1541 either). Both LOAD "$",8 and LOAD "*",8 fail.
Posted By: robiza

Re: SVN builds - new driver flood - 01/23/10 09:52 AM

the first file in the directory it's a fake

you can use:
load"s*",8

run
Posted By: etabeta78

Re: SVN builds - new driver flood - 01/23/10 03:24 PM

a huge thanks to Curt for taking my broken d64 code and turning it into a fully working support for CBM floppy drives
Posted By: Kale

Re: SVN builds - new driver flood - 01/23/10 04:35 PM

Sango Fighter on Super A'Can

http://mamedev.emulab.it/kale/?p=949

Looks playable, and, as a bonus, it has an option mode on which I can finally test & add all the buttons wink
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/23/10 07:33 PM

Originally Posted By robiza
the first file in the directory it's a fake

you can use:
load"s*",8

run


Doesn't work - file not found, and "run" does nothing.
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/23/10 07:35 PM

Originally Posted By Kale
Sango Fighter on Super A'Can

http://mamedev.emulab.it/kale/?p=949

Looks playable, and, as a bonus, it has an option mode on which I can finally test & add all the buttons wink


All buttons are supported. smile

Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/23/10 07:59 PM

Originally Posted By R. Belmont
Originally Posted By robiza
the first file in the directory it's a fake

you can use:
load"s*",8

run


Doesn't work - file not found, and "run" does nothing.


Try to convert .D64 in .G64 format and load the disk with: LOAD "*",8,1

Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/23/10 08:05 PM

How do I convert?
Posted By: Anna Wu

Re: SVN builds - new driver flood - 01/23/10 08:12 PM

Originally Posted By R. Belmont
How do I convert?


You can use the nibble tools or 64Copy

Example (nibble tools) :
nibconv filename.d64 filename.g64
Posted By: Just Desserts

Re: SVN builds - new driver flood - 01/23/10 08:34 PM

Originally Posted By Anna Wu
Originally Posted By R. Belmont
How do I convert?


You can use the nibble tools or 64Copy

Example (nibble tools) :
nibconv filename.d64 filename.g64


(psst... R. Belmont runs Linux)
Posted By: Vas Crabb

Re: SVN builds - new driver flood - 01/23/10 09:45 PM

Hehe - from the end of this thread, it looks like Anna is the master and ArBee is the n00b.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/23/10 09:47 PM

Meh, I planes-trainsed-and-automobiles'ed it and got converted images. Those do load (should I file a bug on the D64 support?) but it locks up after the intro and there's some notable graphics glitches. Still, a year ago it probably wouldn't have run at all so that's progress smile

Also, the arrow keys don't work in natural mode - I had to turn off natural and find a picture of a C64 keyboard to figure out how to select "SID" as the soundcard so it would start up =)

And our SID emulation will shatter windows at 50 meters, but that's leftover untouched PeTcode so I won't complain much there ;-)
Posted By: Duke

Re: SVN builds - new driver flood - 01/23/10 09:47 PM

The nibtools are source available and run on Linux.
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/23/10 09:49 PM

There's no source in that directory and no indication they're portable.
Posted By: Duke

Re: SVN builds - new driver flood - 01/23/10 09:50 PM

http://c64preservation.com/dp.php?pg=nibtools
Posted By: Curt Coder

Re: SVN builds - new driver flood - 01/23/10 10:35 PM

Originally Posted By R. Belmont
(should I file a bug on the D64 support?)


Please do, I'll look into it.
Posted By: Justin

Re: SVN builds - new driver flood - 01/23/10 10:43 PM

Originally Posted By R. Belmont
Also, the arrow keys don't work in natural mode - I had to turn off natural and find a picture of a C64 keyboard to figure out how to select "SID" as the soundcard so it would start up =)


I just tried and I could select SID using the arrow keys in natural keyboard mode on Windows, maybe it's an SDL thing?
Posted By: R. Belmont

Re: SVN builds - new driver flood - 01/23/10 11:04 PM

Works on SDL with other drivers (just tried apple2gs in -natural and all 4 arrows worked).
Posted By: robiza

Re: SVN builds - new driver flood - 01/24/10 07:32 AM

strange!

here my snaps
http://mamedev.emulab.it/robiza/wp-content/uploads/2010/01/0025.png

http://mamedev.emulab.it/robiza/wp-content/uploads/2010/01/0026.png

http://mamedev.emulab.it/robiza/wp-content/uploads/2010/01/0027.png

http://mamedev.emulab.it/robiza/wp-content/uploads/2010/01/0028.png

maybe a bug only with linux os?

edit:
if i use the messui i can't use the d64 files
i need to use mess, tab menu, and select the file
Posted By: Robbbert

Re: SVN builds - new driver flood - 01/24/10 08:52 AM

I just tried MESSUI, the d64 files appeared in the picker, I double-clicked one of them (Space Invaders), and it loaded just fine.
Posted By: robiza

Re: SVN builds - new driver flood - 01/24/10 09:25 AM

with latest release i can't use the messui for .d64 disk

with a clean compile

make TARGET=MESS OSD=winui

src/osd/windows/video.c:71:18: error: menu.h: No such file or directory
src/osd/windows/video.c: In function 'void check_osd_inputs(running_machine*)':
src/osd/windows/video.c:388: error: 'win_toggle_menubar' was not declared in thi
s scope
make: *** [obj/winui/MESS/osd/windows/video.o] Error 1
Posted By: Robbbert

Re: SVN builds - new driver flood - 01/24/10 10:48 AM

Try

make TARGET=mess MESSUI=1

Posted By: robiza

Re: SVN builds - new driver flood - 01/24/10 11:21 AM

thanks
Posted By: