Previous Thread
Next Thread
Print Thread
Page 14 of 39 1 2 12 13 14 15 16 38 39
Joined: May 2012
Posts: 541
S
Senior Member
OP Offline
Senior Member
S
Joined: May 2012
Posts: 541
At the moment ec1841 fails system board test (on the tps2214 floppy from softlist) -- it tests that DMA channel 0 counter is increasing and it doesn't.

Apparently this is because BIOS programs PIT first, DMAC second, and then resets DMAC [1]; but DMAC reset does not clear m_u73_q2 and further DMA cycles aren't triggered.

[1] code starting on line 4880 in https://github.com/shattered/retro-bios/blob/master/ec1841/ec1841.lst

Joined: Mar 2001
Posts: 16,650
Likes: 1
R
Very Senior Member
Online Content
Very Senior Member
R
Joined: Mar 2001
Posts: 16,650
Likes: 1
That's nice work, as usual. AFAIK the DMAC implementation is still pretty new so it's quite possible you've found an edge case.

ETA: both patches are in trunk now.

Last edited by R. Belmont; 09/29/13 03:34 PM.
Joined: May 2012
Posts: 541
S
Senior Member
OP Offline
Senior Member
S
Joined: May 2012
Posts: 541
ec1841 is somewhat easier to work on since there's a technical manual and BIOS listing (but no schematics).

Speech synthesis card for this machine was finally found, complete with TM and schematics -- this one would be fun to emulate. Not much is known about it, but it's possible that it uses a TMS5100 clone: http://www.155la3.ru/k1803.htm

Joined: May 2012
Posts: 541
S
Senior Member
OP Offline
Senior Member
S
Joined: May 2012
Posts: 541
iskr1030m keyboard uses MCS48's ENT0 CLK opcode, which is unimplemented in MESS -- any special reason why it was left out?

Joined: Apr 2004
Posts: 1,554
J
Very Senior Member
Offline
Very Senior Member
J
Joined: Apr 2004
Posts: 1,554
Lack of test cases, I assume.

Joined: Mar 2001
Posts: 16,650
Likes: 1
R
Very Senior Member
Online Content
Very Senior Member
R
Joined: Mar 2001
Posts: 16,650
Likes: 1
Most MAME/MESS CPU cores have opcodes implemented on an as-encountered-in-real-software basis. That way you know when you implement an opcode that you have an immediate test case for it.

Joined: May 2012
Posts: 541
S
Senior Member
OP Offline
Senior Member
S
Joined: May 2012
Posts: 541
OK, so I guess I'll have to fix it :-)

Another thing -- I'm thinking of writing a driver for https://en.wikipedia.org/wiki/DVK and it's early revisions were just a SBC connected via serial port to a terminal (ie15). How do I connect two drivers this way?

Joined: May 2012
Posts: 541
S
Senior Member
OP Offline
Senior Member
S
Joined: May 2012
Posts: 541
i86 cpu core doesn't support some of the 8087 instructions (FINIT, FSTCW, ...) and FPU test in ec1841 fails.

Joined: Feb 2005
Posts: 449
C
Senior Member
Offline
Senior Member
C
Joined: Feb 2005
Posts: 449
Originally Posted By shattered
iskr1030m keyboard uses MCS48's ENT0 CLK opcode, which is unimplemented in MESS -- any special reason why it was left out?


It enables output of system clock on T0 pin which would totally kill performance if a callback handler was attached to it. If it's wired to another CPU's clock input, just use the same clock in the machine config.

Joined: Jan 2012
Posts: 850
C
Senior Member
Offline
Senior Member
C
Joined: Jan 2012
Posts: 850
No FPU in the 8086 currently. A proper implementation would require a bit of work.

Page 14 of 39 1 2 12 13 14 15 16 38 39

Link Copied to Clipboard
Who's Online Now
2 members (Fake Shemp, 1 invisible), 18 guests, and 3 robots.
Key: Admin, Global Mod, Mod
ShoutChat
Comment Guidelines: Do post respectful and insightful comments. Don't flame, hate, spam.
Forum Statistics
Forums9
Topics8,853
Posts116,508
Members4,927
Most Online890
Jan 17th, 2020
Powered by UBB.threads™ PHP Forum Software 7.7.5