| ID | Category [?] | Severity [?] | Reproducibility | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 06740 | Sound | Major | Random | Oct 29, 2017, 18:25 | Mar 11, 2021, 20:23 |
| Tester | Kale | View Status | Public | Platform | MAME (Official Binary) |
| Assigned To | aaron | Resolution | Fixed | OS | Windows Vista/7/8 (64-bit) |
| Status [?] | Resolved | Driver | |||
| Version | 0.191 | Fixed in Version | 0.230 | Build | 64-bit |
| Fixed in Git Commit | 95869d5 | Github Pull Request # | |||
| Summary | 06740: slyspy, bouldash: Sound randomly speeds up and eventually dies | ||||
| Description |
In both these games sound have issues on certain conditions (same sound HW) In Sly Spy, between stage 7 and 8 transitions sound starts to oddly speed up, and might eventually lock up completely with a stuck tone. In Boulder Dash this can occur between stages 2-3 and 2-4 |
||||
| Steps To Reproduce | Play Sly Spy up to stage 8 or Boulder Dash up to 2-4 | ||||
| Additional Information |
Regression version around 0.106 and 0.121, where HuC6280 sound CPU has been downclocked from 3 to 2 MHz. Also provided inps: For Sly Spy WolfMAME 0.190 is used on secretagj set. For Boulder Dash WolfMAME 0.191 is used on bouldashj set, I've continued on 2-4 just to make the bug to occur. |
||||
| Github Commit | |||||
| Flags | |||||
| Regression Version | |||||
| Affected Sets / Systems | slyspy, bouldash | ||||
|
Attached Files
|
 bouldashj_soundbug_wolf0191.zip (125,681 bytes) Oct 29, 2017, 18:25 Uploaded by Kale | ||||
| secretagj_soundbug_wolf190.zip (438,606 bytes) Oct 29, 2017, 18:29 Uploaded by Kale | |||||
 No.14361
Tafoid Administrator
Oct 29, 2017, 23:44
|
Reminds me of Crude Buster where certain sound effects caused the sound to speed up and then go back slow 00389. Might be related somehow? |
|---|---|
 No.14366
Kale Developer
Oct 30, 2017, 10:24
|
Good catch, both shares same CPUs and an IRQ tied to a Yamaha sound chip. There must be something in the CPU execution that makes it to behave strangely ... |
 No.14995
StHiryu Tester
Apr 30, 2018, 23:09
|
Sly Spy speed up sound bug has been fixed in mame 0.192. |
|
No.14998
Kale Developer
May 3, 2018, 03:43
|
Actually worked around by changing the sound CPU clock, not fixed. Changing this to acknowledged. |
 No.16877
MetalGod Senior Tester
Sep 9, 2019, 18:19
|
Is this fixed too ? |
|
No.16878
StHiryu Tester
Sep 9, 2019, 20:11
|
I've just played it at mame 0.212 and the music plays fine, but I don't know if it's because of the cpu clock changes made. I'll try it again in the next git release, since the pull request from the other data east games have been merged into the mame code. |
|
No.16879
MetalGod Senior Tester
Sep 9, 2019, 21:14
|
I mean fixed by recent AJW work, like the related reports. |
 No.16909
AJR Developer
Sep 14, 2019, 16:27
|
My recent changes don't apply to these games, or at least haven't been applied yet. The issue is that some Data East sound programs normally run using IRQs from the YM2151 which override those from the HuC6280 timer (which is reloaded each time by the service routine), but when certain sounds are playing the YM2151 IRQs aren't used for some reason and the music speed is determined by the internally generated IRQs as a result. I have no idea whether something like that is going on with these games as well. |
|
No.18506
StHiryu Tester
Mar 2, 2021, 13:24
|
Sly spy transition to stage 8 works fine after Aaron audio changes. Anyone could test Boulder Dash? |
|
No.18529
MetalGod Senior Tester
Mar 11, 2021, 19:10
|
This is the Boulder Dash bug in MAME 0.191. Very easy to reproduce: |
|
No.18530
MetalGod Senior Tester
Mar 11, 2021, 19:54
edited on: Mar 11, 2021, 19:54
|
I'm unable to reproduce the bug in boulder dash in current mame I'm closing and resolving this bug |