Previous Thread
Next Thread
Print Thread
Page 1 of 9 1 2 3 4 5 6 7 8 9
#21824 07/06/05 03:49 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
http://www.bannister.org/software/ao.htm

It's now available in a 4th flavor: native 64-bit Linux. So AO can now play your favorite 8-bit music on your new 64-bit computer :-)

And everyone gets these improvements:
* New, far better sounding engines for NSF, GBS, and SNDH. The YM2149 never sounded so good ;-)
* CAB archive support so you can play S98s from Personal Soundlog Page directly
* Improvements to MOD and S3M playback
* Fixes for crashes with certain RAR and RAW files
* QSFs play at the right speed now
* S98s don't click and many more of them play correctly and at the right tempo

#21825 07/06/05 04:08 AM
Joined: Dec 1969
Posts: 920
Likes: 3
R
Senior Member
Senior Member
R Offline
Joined: Dec 1969
Posts: 920
Likes: 3
S98s still click from time to time; just less often than before. Try the Thexder Moonlight Sonata one; it'll click at some stage during playback. It's not always in the same place, suggesting a memory initialisation problem of some kind.

#21826 07/06/05 04:19 AM
Joined: Dec 1969
Posts: 920
Likes: 3
R
Senior Member
Senior Member
R Offline
Joined: Dec 1969
Posts: 920
Likes: 3
I've found the problem. Working on it now.

#21827 07/06/05 04:33 AM
Joined: Dec 1969
Posts: 920
Likes: 3
R
Senior Member
Senior Member
R Offline
Joined: Dec 1969
Posts: 920
Likes: 3
Right. S98 is improved in 2.0b6, but will be even better in 2.0b7 now that the second sound chip is updated properly smile

#21828 07/06/05 04:53 AM
Joined: Nov 2000
Posts: 114
V
Senior Member
Senior Member
V Offline
Joined: Nov 2000
Posts: 114
PSF tempo is still too fast, though it doesn't look like that was worked on so I suppose that's to be expected.

NSF files sound outstanding now though. Great update.

#21829 07/06/05 06:12 AM
Joined: May 2005
Posts: 62
S
Member
Member
S Offline
Joined: May 2005
Posts: 62
the SNDH plays much better now laugh
but they sure sound a bit too muffled?
here is a zip with 3 recordings from real Atari ST , also with the SNDH files of the tunes... the worst example is the Kamikaze tune.

And i am only able to listen to 3 minutes of the sndh files in AO (windows) ?

#21830 07/06/05 06:38 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
I don't impose any time limit, but perhaps SC68 is expecting some tag format and defaulting to 3 minutes. I'll look into it.

#21831 07/06/05 06:53 AM
Joined: Jun 2005
Posts: 16
Junior Member
Junior Member
Joined: Jun 2005
Posts: 16
Yes SNDH and GBS sound a LOT better! �Ķwill check other formats when i'm less tired smile

However�Ķ there is now the rather 'fun' bug with the SNDH files. Files that contain multiple songs now play as one continous track. The 'fast forward' button does not work as intended (it actually tries to skip the tracks backwards into minus numbers) and the 'rewind' button is constantly greyed-out.

'Fun' because with the absence of playlists it is actually quite nice to not have to manually select each track ^^.

I'm also having the same problem as Stefan with the 3 minutes-and-stop. (but not, I don't think with the files with multiple tracks).

Anyways, thanks for a swift update!

Mac OS X Version

#21832 07/06/05 07:25 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
I don't think I have any multi-song sndh files - could you suggest one or two that do that? smile

#21833 07/06/05 07:43 AM
Joined: May 2005
Posts: 62
S
Member
Member
S Offline
Joined: May 2005
Posts: 62
in the zip i linked in my previous post there is a multitune SNDH the kamikaze one :-)

#21834 07/06/05 08:06 AM
Joined: Oct 2004
Posts: 296
S
Senior Member
Senior Member
S Offline
Joined: Oct 2004
Posts: 296
There are a few NSF files that cause an error dialog to appear in OS X, the kind that says it's either a nested file or unsupported. After this dialog, the emulator crashes. The NSF files affected in 2.0b6 are: (might be more)
-Die Hard
-Doki Doki Panic
-Indiana Jones and the Last Crusade (one of the versions; the other plays fine)
-Mr. Gimmick!
-Power Blade

Otherwise, good release. I like how the GBS files actually play at actual speed now instead of how slow they were before. To be honest, I couldn't really tell much of a difference between the NSF playback in 2.0b5 and 2.0b6, except that Kirby's Adventure now sounds correct. Thanks for that.

#21835 07/06/05 10:00 AM
Joined: Feb 2004
Posts: 312
A
Senior Member
Senior Member
A Offline
Joined: Feb 2004
Posts: 312
Quote
Originally posted by Richard Bannister:
Right. S98 is improved in 2.0b6, but will be even better in 2.0b7 now that the second sound chip is updated properly smile
Great! It is better in 2.0b6 but some of the instruments are still really wrong. I can host a track or three from "Music From Sorcerian" if it'd help to hear the originals.

#21836 07/06/05 10:01 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
Gimmick doesn't work because the new NSF engine only supports Namco 106 (N106) and Konami VRC6 expansion sound (it doesn't support VRC7 or MMC5). When I tried Gimmick! and got the "not supported" error, the next NSF file I opened triggered an internal assertion failure on the console. It looks like a minor bug in AO's glue code for the new NSF engine.

eng_nsf.cpp:34: failed assertion `emu == NULL'

I tried the other files from Zophar and found that they incorrectly specify that they use several expansion sound chips, thus cause AO to quit as described above. The expansion sound byte at offset 0x7B in these files should be zero. Kevin Horton\'s NSF archive has rips of these that work fine.

#21837 07/06/05 08:30 PM
Joined: May 2005
Posts: 22
Junior Member
Junior Member
Joined: May 2005
Posts: 22
Quote
Originally posted by Richard Bannister:
Right. S98 is improved in 2.0b6, but will be even better in 2.0b7 now that the second sound chip is updated properly smile
Oh, good! It's already a fair improvement, so I quite look forward to that. Do you know if it fixes the buzzing in Yu-No as well?

#21838 07/06/05 09:48 PM
Joined: May 2002
Posts: 67
N
Member
Member
N Offline
Joined: May 2002
Posts: 67
Cool, a few of the SNDH files I was having trouble with in b5 work great now!

Small HES problem, the bass in songs from Raiden and Soldier Blade gets pretty weird. in_he2.dll and NEZPlug don't exhibit this behavior. I can provide comparison recordings from my TurboGrafx if needed for troubleshooting.

I'm glad to finally have a Windows-based player that runs Eternal Blue Complete PSFs at the correct tempo! They've been slow for me in HE for a long time now... with the correct rip no less.


-nZero
#21839 07/07/05 12:26 AM
Joined: Dec 1969
Posts: 920
Likes: 3
R
Senior Member
Senior Member
R Offline
Joined: Dec 1969
Posts: 920
Likes: 3
Quote
Originally posted by Vlcice:
Do you know if it fixes the buzzing in Yu-No as well?
I only tried two of the tracks at random, but neither have buzzing - so probably.

#21840 07/07/05 12:51 AM
Joined: May 2005
Posts: 22
Junior Member
Junior Member
Joined: May 2005
Posts: 22
Ah, great. :3 Thank you very much!

#21841 07/07/05 04:12 AM
Joined: Oct 2004
Posts: 296
S
Senior Member
Senior Member
S Offline
Joined: Oct 2004
Posts: 296
Blargg, about the correct NSF files: I went to that site you linked to but I couldn't find any downloads. Do I have to have EDonkey to access them?

#21842 07/07/05 09:18 PM
Joined: Dec 1999
Posts: 1,180
Likes: 2
J
Very Senior Member
Very Senior Member
J Offline
Joined: Dec 1999
Posts: 1,180
Likes: 2
gamesnet.raw still crashes (using the Win32 version on an Athlon XP).

ode2ptk.mod is now even worse although I understand if you don't want to listen to it further smile

NSF and GBS sound quite good now, I'll have to do some comparisons to real hardware.

#21843 07/07/05 09:55 PM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
The MOD change fixed a very bad error in U4ia's "Watch Your Bass Bins". I don't see how it would affect ode2ptk.

#21844 07/08/05 03:28 AM
Joined: May 2002
Posts: 67
N
Member
Member
N Offline
Joined: May 2002
Posts: 67
QSF still plays quite a bit slower than the same games in M1, for me. However, I only tested with these sets .

By the way, here's stuff related to troubleshooting the weirdness with HES files/PC Engine stuff. A recording of the first stage of Raiden from my TG16. Compare to the HES for the game in AO. Also, Soldier Blade's 2nd stage ( song 9 in AO ) exhibits severe bass-related oddness, too.

O/T: Why aren't there any HES rips for the chip songs on Super CD-ROM games? frown


-nZero
#21845 07/08/05 03:46 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
Speedy Boris, it looks as if you need emule (or edonkey) to get it now. I've never used those programs. It was probably taking too much bandwidth otherwise.

It's pretty simple to fix them with a hex editor as I described (send me a private message for further help). I think I'll update the sound engine to allow ignoring the sound chip flags (so you can play these incorrect files, as apparently the previous engine did) and get it to the AO team before the next release.

#21846 07/08/05 04:02 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
I don't have those sets, but e.g. 19xx (from Neill Corlett's homepage) was A/B'ed to play at the same speed as M1.

#21847 07/08/05 07:27 PM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Looks like gamesnet would only crash on Win32. I upgraded my MinGW to GCC 3.4.4 (which matches what I build the Linux versions with) and did some other tweeks and it seems to be OK now.

#21848 07/08/05 08:34 PM
Joined: May 2002
Posts: 67
N
Member
Member
N Offline
Joined: May 2002
Posts: 67
Quote
Originally posted by R. Belmont:
I don't have those sets, but e.g. 19xx (from Neill Corlett's homepage) was A/B'ed to play at the same speed as M1.
::shrug:: I don't know what to say, both Power Battles and Power Fighters work fine under Highly Quixotic. Oh well.


-nZero
#21849 07/08/05 11:49 PM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
It's because of how Corlett converts them to work without the 68000 driver (in M1, *you're* the driver) - it needs a faster clock speed than normal to work properly. It'll be fixed in the next version. (Although QSF is the world's most pointless format anyway given the existance of M1).

#21850 07/09/05 01:32 AM
Joined: May 2002
Posts: 67
N
Member
Member
N Offline
Joined: May 2002
Posts: 67
Quote
Originally posted by R. Belmont:
(Although QSF is the world's most pointless format anyway given the existance of M1).
Agreed. Thanks for the explanation. Anyway, until the next M1 is out AO is the only way I can listen to the QSound version of Power Battle without pointless added resampling, so I appreciate your looking into it smile


-nZero
#21851 07/09/05 09:48 AM
Joined: Jun 2000
Posts: 23
Member
Member
Joined: Jun 2000
Posts: 23
Quote
Originally posted by R. Belmont:
I don't think I have any multi-song sndh files - could you suggest one or two that do that? smile
The SNDHs for Hammerfist and Reach For The Skies are multi-tune files as well.

I'm also experiencing NSF crash problems- I'll try to play a tune that causes an error, then the next NSF i play will crash the program.

Also, there's several NSFs that use more than one expansion chip. This is impossible on real hardware, but several do it anyway. Any plans for supporting this?

#21852 07/09/05 10:10 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
I couldn't find any NSFs that used VRC6 and N106 expansion sound simultaneously, so I wasn't able to fully test the engine with multiple active sound chips. Pointers to any would be appreciated.

I posted earlier about this bug and about fewer expansion sound chips being supported by the new NSF engine.

#21853 07/09/05 07:11 PM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Yes, if you try and play an NSF that causes an error you must shut down the app before attempting to play another.

#21854 07/11/05 10:59 PM
Joined: Dec 1999
Posts: 1,180
Likes: 2
J
Very Senior Member
Very Senior Member
J Offline
Joined: Dec 1999
Posts: 1,180
Likes: 2
Any news on speeding up the win32 oscilloscope, which was "on the todo list" several versions ago? wink

#21855 07/12/05 12:49 PM
Joined: Jun 2000
Posts: 23
Member
Member
Joined: Jun 2000
Posts: 23
Several multi-expansion NSFs can be found in recent Famicompo entries.

#21856 07/13/05 06:49 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
That's cheating. If I ran a compo like that I'd limit it to real-world hardware configurations smile

#21857 07/13/05 08:22 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
I scanned all the NSFs from the Famicompo vol. 1 and vol. 2 and none used only the N106 and VRC6 simultaneously.

#21858 07/15/05 02:58 AM
Joined: Jun 2005
Posts: 16
M
Junior Member
Junior Member
M Offline
Joined: Jun 2005
Posts: 16
Quote
Originally posted by blargg:
Speedy Boris, it looks as if you need emule (or edonkey) to get it now. I've never used those programs. It was probably taking too much bandwidth otherwise.

It's pretty simple to fix them with a hex editor as I described (send me a private message for further help). I think I'll update the sound engine to allow ignoring the sound chip flags (so you can play these incorrect files, as apparently the previous engine did) and get it to the AO team before the next release.
At least post the trouble ones.

#21859 07/16/05 04:12 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
I made a zip archive of Die Hard, Doki Doki Yuuenchi, Indiana Jones and the Last Crusade, and Power Blade. If someone contacts me with a place to host the entire 4MB RAR of Kevin Horton's NSF rips, I'll provide it and post a link. I don't have 4MB of web space and bandwidth to host it myself.

#21860 07/16/05 04:22 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
4MB? I imagine Richard can handle that.

#21861 07/16/05 09:00 PM
Joined: Oct 2004
Posts: 296
S
Senior Member
Senior Member
S Offline
Joined: Oct 2004
Posts: 296
Erm, you included the wrong "Doki Doki" title there, blargg. The one that crashed AO for me was Doki Doki Panic (the precursor to Super Mario 2), not Yuuenchi. Then again, maybe they BOTH don't work. I dunno. Thanks for the others, though!

#21862 07/17/05 05:00 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
That's the only "Doki Doki" I had. I figured it might be the one. ... I just researched it and Doki Doki Panic uses Famicom Disk System sound, which isn't supported by the new NSF sound engine.

Also I just want to point out that AO isn't crashing, rather an assertion (check for possibe bug) inside is failing which is quitting the program immediately. Basically there's a part that says "such-and-such should be the case here; if it's not, quit the program immediately". A fine point maybe, but to me a crash is when a program makes a bad memory access or goes haywire internally. Both are due to a bug, but the latter is more serious than the former. And just to avoid misrepresentation, my only involvement with Audio Overload development is my authorship of the NSF and GBS sound engines it now uses; I have not worked on the main program at all.

#21863 07/17/05 10:55 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
The assertion is fixed, incidentally (although not in a public build yet).

BTW, were you planning to support more of the add-on NES hardware like FDS or VRC7?

#21864 07/17/05 01:31 PM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
I probably won't be writing any more expansion sound chip emulators. As I remember, MMC5 and VRC7 are rarely used. I was going to support FDS a while back but couldn't find reliable documentation. Having never even heard the thing, I didn't go further.

I care most about exact emulation; if I can't get very close I won't bother. N106 and VRC6 were simple enough that I implemented them based on documentation alone. GBS sound is still somewhat problematic because I don't have a Game Boy development cartridge for testing the sound hardware. Documentation always falls short in some way and it's not fun going in circles trying to figure out what's supposed to happen in ill-defined cases.

It might be possible to run other FDS, MMC5, and VRC7 emulators in parallel. The simplest solution of course is to include an alternate NSF engine that supports everything, and use it when the expansion sound bits in the NSF header have any bits other than 0x11 set.

As an alternate NSF sound engine with support for (I think) all the expansion sound chips, I'd recommend Disch\'s NotSoFatso , or the band-limited rewrite in his experimental Schpune. I think they're both under the GNU GPL license (I'm e-mailing the author about alternate licensing). I'd even volunteer to separate the NSF playing code from NotSoFatso or Schpune into a more independent player library.

#21865 07/18/05 12:27 AM
Joined: May 2002
Posts: 67
N
Member
Member
N Offline
Joined: May 2002
Posts: 67
I'm probably wrong, but I thought the VRCVII was just an OPLL with additional h/w for memory mapping-- could MAME's YM2413 core be fudged in or something? Not a huge deal, since only one commercial game uses it, but it seems to be somewhat popular with doujin NSFs.


-nZero
#21866 07/18/05 12:59 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
No idea, but if that's the case I could easily hack in Jarek's like you say.

UPDATE: Yeah, Charles Mac Donald says it's a YM2413, and that's good enough for me :-)

#21867 07/19/05 04:50 AM
Joined: Jul 2001
Posts: 105
Likes: 1
R
Senior Member
Senior Member
R Offline
Joined: Jul 2001
Posts: 105
Likes: 1
What about support for the NSFe format? It's just an extension to NSF, but if you try to play an NSFe in AO it just shows gibberish in the song titles and doesn't make any sound.

http://disch.zophar.net/nsfe.php

#21868 07/19/05 06:14 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
It's rather more than "just an extension". That said, support would be nice. blargg, you up for it?

#21869 07/19/05 09:07 PM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
OK, I just rewrote the my player's NSFE code as a standalone module and tested it a bit. It allows use of the following NSFE fields (others can be easily added if necessary):

Game, author, copyright, person who ripped music, names of individual tracks, lengths of individual tracks, and an optional playlist. In most NSFEs the playlist contains the music tracks in the order they occur while playing the game, leaving out the sound effects and short music sequences.

#21870 07/20/05 04:19 AM
Joined: Jul 2005
Posts: 1
M
Junior Member
Junior Member
M Offline
Joined: Jul 2005
Posts: 1
One small issue that's bugging me is that on OS X, the "Export to WAV" dialog box only allows filenames up to 31 characters long. Is there any intention on allowing longer file names in the future?

#21871 07/20/05 11:55 PM
Joined: Dec 1969
Posts: 920
Likes: 3
R
Senior Member
Senior Member
R Offline
Joined: Dec 1969
Posts: 920
Likes: 3
I'll resolve that at some stage, though I can't say that is a priority for me.

Richard in Venice

#21872 07/25/05 04:49 AM
Joined: Jul 2005
Posts: 1
Junior Member
Junior Member
Joined: Jul 2005
Posts: 1
Is there any place to download old builds of AO? There are some strange things happening with some of my SPCs, and I think the previous version I had played them fine... but I want to check.


"...and me, with this terrible pain in all the diodes down my left side..."
#21873 07/26/05 06:40 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
AFAIK there's no archive. Sorry.

#21874 07/27/05 11:05 AM
Joined: Jun 2005
Posts: 16
M
Junior Member
Junior Member
M Offline
Joined: Jun 2005
Posts: 16
Another thing that doesn't work with the new NSF engine is Turtles Tournament Fighters. However, since the other three Turtles soundtracks play fine, it might be a problem with the the .NSF itself.

www.zophar.net/nsf/turtle.zip

#21875 07/28/05 12:20 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
Turtles Tournament Fighters plays fine on on my machine (Mac OS X). It has 46 sound effect tracks before the actual music.

#21876 07/28/05 03:21 AM
Joined: Jun 2005
Posts: 16
M
Junior Member
Junior Member
M Offline
Joined: Jun 2005
Posts: 16
Certain tracks(like number 47)don't play right.

#21877 07/28/05 03:41 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
You sure it's not that other players get it wrong?

#21878 07/28/05 03:46 AM
Joined: Jun 2005
Posts: 16
M
Junior Member
Junior Member
M Offline
Joined: Jun 2005
Posts: 16
It doesn't play right in Audio Overload AND Game Music Box.

#21879 07/28/05 08:16 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Again, you have real hardware and can verify that AO/GMB are wrong and not all the other players? smile

#21880 07/28/05 08:52 AM
Joined: Jan 2005
Posts: 154
Senior Member
Senior Member
Joined: Jan 2005
Posts: 154
I tried TMNT: TF on a local build of Festalon (another quite accurate player engine), and on my experimental NSF player which uses an actual NES for sound output, and they sound the same as AO 2.0b6 (other than slightly different frequency equalizations on each). Here is the sample from playing the NSF through my NES:

tmnt_tf_47.mp3 (320K)

This isn't definitive though, since it's not running the NSF code on the NES CPU (my devcart doesn't have enough RAM to hold the entire NSF), and I don't have the actual game.

So, it could be 1) an error on your part, 2) the players, or 3) a problem with the NSF rip (in order of likelihood).

#21881 07/28/05 01:12 PM
Joined: Nov 2003
Posts: 806
S
smf Offline
Senior Member
Senior Member
S Offline
Joined: Nov 2003
Posts: 806
Quote
Originally posted by blargg:
So, it could be 1) an error on your part, 2) the players, or 3) a problem with the NSF rip (in order of likelihood).
Bad rips that work in bad players make that a difficult order to guess. There is definately one of the PSF's that fall into that category, I haven't verified the other suspect ones on real hardware yet ( mainly because I don't know exactly which they are :-) ).

smf

#21882 08/17/05 11:11 PM
Joined: Aug 2005
Posts: 21
T
Junior Member
Junior Member
T Offline
Joined: Aug 2005
Posts: 21
8000 clip incidents happen in an average snd export file right now. i will test other formats and post results

i agree with Stefan Lindberg: the snd emulation does sound very compressed. But i will say that before i heard the real atari examples i still enjoyed the compressed version. ( although it was not true to the original sound) smile

and i have read on 1 site that chocobo dungeon 2 was ripped incorrectly. (a japanese site that was very upset about it)


the times they are a changin
#21883 08/18/05 02:48 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Yes, CD2 is an incorrect rip - it has an illegal MIPS code sequence (as was well documented on smf's and my blogs). It works anyway in Highly Experimental (the "reference" PSF/PSF2 player) because HE's CPU is buggy. It works in AO because we detect the bad code and do what it means rather than what it says.

And by modern pop/rock CD mastering standards, 8000 clips in a single song is average ;-) That said, please specify the actual format and song used to compile that stat.

#21884 08/18/05 06:22 AM
Joined: Aug 2005
Posts: 21
T
Junior Member
Junior Member
T Offline
Joined: Aug 2005
Posts: 21
i am using elemental audio InspectorXL. And all snd files in 2.0b6 are cut to 3min by default and the audio output is clipping to begin with. that could be the cause of some of the sound compression. Anyways i will try to do some more tests on different formats to see if they are clipping the output as well.

This was without normalization or optimizing sound functions.
Normalizing is only for low level signals anyways. this is too high and when normalized, gets even more compressed.

And if you are talking about 8000 clips being normal: for tape yes, for digital NONONO

Also the songs specifically were from YMrockers and DropDaBomb snd files but other snd files acted the same way with the normalize function off or on


the times they are a changin
#21885 08/18/05 06:56 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
It's completely normal for major label CDs to clip frequently since 1998. Modern ProTools compression plugins let you create full-on square waves that don't sound bad to the untrained ear - you'll just notice your ears become "fatigued" after a while.

(Ironically, this means most CDs are already shipped at such a bad quality now that MP3 can't do any further damage to them).

#21886 08/18/05 08:12 AM
Joined: Apr 2005
Posts: 29
Member
Member
Joined: Apr 2005
Posts: 29
PSFs aren't good for slow computers... especially for my 300 mhz G3!


KungFuFurby, Bug Reporter for various stuff (somewhat)

Runs: Mac OS X 10.4.11 on a 1.8 GHz PPC G5
#21887 08/18/05 11:21 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Well, yes, playing them involves emulating most of a PS1 except the graphics :-) (And some even require the GPU status bits to at least show Vblank properly - ugh).

#21888 08/18/05 06:28 PM
Joined: Aug 2005
Posts: 21
T
Junior Member
Junior Member
T Offline
Joined: Aug 2005
Posts: 21
re: R belmont. Yes, I know I am a mastering engineer and often get paid to fix gain stages in protools, logic, ableton, acid,etc.. mixes before i master because good mastering only accentuates the audio and as a result the defects as well. But i sure hope this isn't a "2 wrongs make a right" analysis.;p

and yes, i have hyperaccusis (increased sensitivity to air pressure changes of all kinds, due to the basila fluid in my choclea existing in higher levels than normal) so fatigue of the ears and consequent headaches is a big avoidance priority for me. But the ear itself is the most sensitive part of your body and is easily damaged for anyone. It is also one of the more mysterious parts of the body due to it's decomposition rate of less than a few hours after death. so most of the world is still "in the dark" about the ear.

Also the clipping (overdriving of the signal) adds to the un-intentional compression of the sound.


the times they are a changin
#21889 08/19/05 01:13 AM
Joined: Apr 2005
Posts: 29
Member
Member
Joined: Apr 2005
Posts: 29
We've got a problem...

In SuperMarioBrothers3.nsf, some songs play with the kickdrum repeating very quickly. Oh dear...


KungFuFurby, Bug Reporter for various stuff (somewhat)

Runs: Mac OS X 10.4.11 on a 1.8 GHz PPC G5
#21890 08/19/05 03:18 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
techne: no, I'm not at all claiming that we should abandon audio quality even though the RIAA has :-)

AO applies no gain stages to the output of the ST emulation though, assuming normalization is off. Perhaps I'm misconfigurating SC68 somehow.

#21891 08/19/05 09:42 AM
Joined: Aug 2005
Posts: 21
T
Junior Member
Junior Member
T Offline
Joined: Aug 2005
Posts: 21
i am a bit busy but in the next week i can thoroughly analyze each supported platform (except QSF since i have none) and list songs used and and a whole slew of diagnostic information that may be helpful in applying precise changes to any of the formats gain implementations. Anyways since i take it you are an engineer too, i thought you might be interested.


the times they are a changin
#21892 09/04/05 01:42 PM
Joined: Sep 2005
Posts: 1
P
Junior Member
Junior Member
P Offline
Joined: Sep 2005
Posts: 1
It seems maybe there are some timing problems with the PSF format? Have only tried the rip of Xenogears from Zophar's Domain, using a G4 1.67 ghz Powerbook/ OSX . For instance, the track "Tams, the Man of the Sea" plays fast vs. winamp/HE (PC/XP) and my pretty well burned in recall of the actual PSX track. Sounds great, but it's the wrong tempol wink This seems to be the case for the whole PSF in my limited testing.

#21893 09/18/05 08:22 PM
Joined: Apr 2005
Posts: 29
Member
Member
Joined: Apr 2005
Posts: 29
Yes, I feel something's weird with the tempo on some songs with PSF files. Aren't we lucky that PSFPlay is now not the only Mac Only PSF player?


KungFuFurby, Bug Reporter for various stuff (somewhat)

Runs: Mac OS X 10.4.11 on a 1.8 GHz PPC G5
#21894 09/19/05 02:21 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
PSF's a hair fast. I didn't think it would matter that much (it's less than a 5 BPM difference) but apparently everyone's a perfectionist about the Squaresoft crap.

-RB far prefers Konami soundtracks

#21895 09/19/05 04:36 AM
Joined: Oct 2004
Posts: 296
S
Senior Member
Senior Member
S Offline
Joined: Oct 2004
Posts: 296
I'd rather soundtracks be too fast than too slow, myself. wink

#21896 09/19/05 08:16 AM
Joined: Nov 2000
Posts: 114
V
Senior Member
Senior Member
V Offline
Joined: Nov 2000
Posts: 114
Actuate emulation, that's all we're after! wink

I think everyone is just pointing out the bug, is all.

#21897 10/11/05 04:46 AM
Joined: Mar 2005
Posts: 42
I
Member
Member
I Offline
Joined: Mar 2005
Posts: 42
I suppose it has been established that VRCVII files (well, the one file) cause AO to crash. That's too bad, I love listening to Lagrange point.
However, on another note, the VGM code is buggy in its emulation of periodic noise. The current noise is entirely untuned, which makes listening to such tunes (Matt Furniss makes a lot of use of periodic noise, as do current doujin composers) sounds like a really bad SAP.

#21898 10/11/05 06:09 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
It really shouldn't crash (I think we actually have it fixed so it just doesn't play now), but VRCVII support is kinda up to blargg at this point.

AFAIK the VGM support is using the latest MAME cores, and I know for the YMxxxx at least that Jarek has verified those sample-per-sample against the real chips, including the noise. The PSG for the Master System may not be so lucky though.

#21899 10/11/05 05:30 PM
Joined: Dec 1969
Posts: 920
Likes: 3
R
Senior Member
Senior Member
R Offline
Joined: Dec 1969
Posts: 920
Likes: 3
VGM support *is* using the latest MAME cores. In theory I could switch to blargg's implementation, but right now that doesn't support the FM chip so I'm not going to do that.

#21900 10/12/05 07:18 AM
Joined: Aug 2003
Posts: 122
T
Senior Member
Senior Member
T Offline
Joined: Aug 2003
Posts: 122
I'm sure this isn't the right place to ask, but where is -eh? So why not?

Other than hunting down an OST of MegaMan-X4; I did find a site that does have some MP3s (actually when you try to save them all you get is a web-page 'html'). They have 4 MP3s. 2 of them you can already listen to on the disc. And the other 2 are said to have been ripped from the data tracks on the disc.

I've even looked around over at Zophar's. Nothing there. Will AO ever support whatever file types are in Saturn discs?

Or perhaps someone knows of a way to convert the data on a Saturn game disc to something that's listenable... :hmm:

Thanks for any info.


Dear Mod/Admin, Please let me know why you had to delete this post. Thanks.
#21901 10/12/05 07:45 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Unlike the PSX the Saturn didn't really have any standard audio formats - they pretty much were different per developer and sometimes per game.

#21902 10/12/05 08:44 AM
Joined: Aug 2003
Posts: 122
T
Senior Member
Senior Member
T Offline
Joined: Aug 2003
Posts: 122
Hmm... Now I'm reminded of what someone said over at mame.net. About the STV not having any 'standards'. lol

If the link in this post isn't allowed then (of course) delete it (and accept my apologies). This is what I was talking about when I mentioned the 2 tracks that were ripped from the 'data' files of the CD... That is: if they can even be DL'ed.

I wonder how they did it? :hmm:

segaxtreme


Dear Mod/Admin, Please let me know why you had to delete this post. Thanks.
#21903 10/12/05 07:16 PM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Someone figured out the format Capcom used for those tracks, apparently. That doesn't mean it's a universal solution for all the tracks in that game though, let alone all ST-V/Saturn games smile

#21904 10/12/05 11:04 PM
Joined: Oct 2002
Posts: 121
L
Member
Member
L Offline
Joined: Oct 2002
Posts: 121
Quote
Originally posted by R. Belmont:
Someone figured out the format Capcom used for those tracks, apparently. That doesn't mean it's a universal solution for all the tracks in that game though, let alone all ST-V/Saturn games smile
...'for those tracks'? What kind of evil thing did Sega made?!?

#21905 10/13/05 04:04 AM
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Nothing, they just gave developers free reign. Like a PC :-)

#21906 11/04/05 09:27 PM
Joined: Oct 2004
Posts: 296
S
Senior Member
Senior Member
S Offline
Joined: Oct 2004
Posts: 296
Quote
Originally posted by Furby:
We've got a problem...

In SuperMarioBrothers3.nsf, some songs play with the kickdrum repeating very quickly. Oh dear...
I know this reply is VERY late, but if you're still reading this, I may have a solution...

There are TWO Super Mario 3 sets at zophar.net, one with a copyright of 1990, and another of 19??. The 1990 one plays fine for some reason, and lo and behold, I tried the other set (the 19??), and it DOES sound really bad. Since they both contain the same tracks, you might as well just use the 1990 version. smile

Page 1 of 9 1 2 3 4 5 6 7 8 9

Moderated by  R. Belmont, Richard Bannister 

Link Copied to Clipboard
Who's Online Now
4 members (Darkstar, farngle, hal3000, 1 invisible), 58 guests, and 2 robots.
Key: Admin, Global Mod, Mod
ShoutChat
Comment Guidelines: Do post respectful and insightful comments. Don't flame, hate, spam.
Forum Statistics
Forums9
Topics9,331
Posts122,197
Members5,077
Most Online1,283
Dec 21st, 2022
Our Sponsor
These forums are sponsored by Superior Solitaire, an ad-free card game collection for macOS and iOS. Download it today!

Superior Solitaire
Powered by UBB.threads™ PHP Forum Software 8.0.0