Viewing Issue Advanced Details
ID Category [?] Severity [?] Reproducibility Date Submitted Last Update
06682 Sound Minor Always Sep 11, 2017, 22:29 Sep 15, 2017, 01:09
Tester M.A.S.H. View Status Public Platform MAME (Official Binary)
Assigned To AJR Resolution Fixed OS Windows 10 (64-bit)
Status [?] Resolved Driver
Version 0.189 Fixed in Version 0.190 Build 64-bit
Fixed in Git Commit Github Pull Request #
Summary 06682: qb3: Missing AY-3-8910 sound
Description Missing sound in QB-3 (prototype) since MAME 0.189.
QB-3 uses 3x AY-3-8910 for sound, like Demon (same driver). But in Demon the sound works!
Steps To Reproduce
Additional Information Changes in 0.189:
- 0.189: Added new 74LS259 device to cinemat.cpp [AJR].
see https://git.redump.net/mame/commit/?id=dab683e78f0bed3f786a9206d905027c1ffba9a6
--> audio/cinemat.cpp
--> drivers/cinemat.cpp
Github Commit
Flags
Regression Version 0.189
Affected Sets / Systems qb3
Attached Files
 
Relationships
There are no relationship linked to this issue.
Notes
3
User avatar
No.14187
AJR
Developer
Sep 14, 2017, 21:33
edited on: Sep 14, 2017, 21:39
This is not going to be a fun job. QB-3 is an unreleased prototype by Rock-Ola using an additional Z80-based sound board similar to the one they added to Demon (a schematic for which is available), but with the communications modified in some undocumented way that wasn't perfectly emulated before I broke it. It really doesn't help here that the Cinematronics CPU isn't an actual integrated circuit but a loose assortment of TTL parts, which makes it ambiguous as to what data the output instructions are supposed to be writing into the FIFO.
User avatar
No.14188
M.A.S.H.
Senior Tester
Sep 14, 2017, 23:09
Or you revert the changes only for QB-3 and added a note.
User avatar
No.14189
AJR
Developer
Sep 15, 2017, 01:09
Simply reverting to the old code for just one machine wouldn't have worked, given that the old qb3 implementation was too closely tied to demon and other games in the driver. It took a temporary reversion before I saw that the old implementation was inverting the sound command data once in the qb3-specific routine, only to invert it yet again in the one used by demon. I've dealt with that for now.