Viewing Issue Advanced Details
ID Category [?] Severity [?] Reproducibility Date Submitted Last Update
01622 Interface Major Always Mar 25, 2008, 15:15 Apr 4, 2008, 13:33
Tester Mr_Person View Status Public Platform MAME (Official Binary)
Assigned To aaron Resolution Fixed OS Windows XP/Vista 64-bit
Status [?] Resolved Driver
Version 0.124a Fixed in Version 0.124u2 Build 64-bit
Summary 01622: -refreshspeed has no effect
Description Even with -refreshspeed enabled, running a game with a higher refresh rate than the host system's monitor (For example, pacman, 60.6Hz, on a 60Hz LCD.) causes the usual video issues. Speed is 100% when it clearly shouldn't be.

Previous official binaries have -refreshspeed working properly on the same PC.
Steps To Reproduce
Additional Information
Flags
Regression Version
Affected Sets / Systems
Attached Files
 
Relationships
There are no relationship linked to this issue.
Notes
5
User avatar
No.00407
Tafoid
Administrator
Mar 25, 2008, 20:40
As I understand the command, -refreshspeed is only used when using a video syncing command (-syncrefresh, -triplebuffer, -waitvsync). My tests here (pacman, mrdo, phoenix) show that it does indeed work when using those commands and stays at a solid 100% when not. To me, it makes little sense to slow down emulation to match monitor refresh speed if you are not going to sync it to this lower speed. Ensure you have a syncing command on and use -autoframeskip to alleviate any stutters that might occur.

Any other comments?
User avatar
No.00408
Mr_Person
Tester
Mar 25, 2008, 20:59
Running, say, "vmame64 pacman -refreshspeed -triplebuffer" on 0.123 results in perfectly smooth audio and video. The same command on 0.124a results in audio pops and occasional jerkiness, as if it were still running at 60.6Hz.

I didn't bother trying -refreshspeed without a syncing command because I assumed it's intentionally ignored there, as seems to be the case.
User avatar
No.00409
aaronmod
Developer
Mar 25, 2008, 21:42
I'm not surprised it is broken. The speed parameter is computed before the system is launched, and the way we fetch screen refresh information at that point changed. I don't think we updated the code there.
User avatar
No.00552
aaron
Developer
Apr 4, 2008, 13:32
Actually, it's broken. I didn't fix it all the way.
User avatar
No.00553
aaron
Developer
Apr 4, 2008, 13:32
Ok, now it's fixed for real.