https://bugs.winehq.org/show_bug.cgi?id=41329
Bug ID: 41329 Summary: Regression in dsound latency (pulse) Product: Wine Version: 1.9.18 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: lolisamurai@tfwno.gf Distribution: ---
As for my other bug report, this report is specific to the game osu! which can be installed with the same steps as the older bug report (https://bugs.winehq.org/show_bug.cgi?id=41007).
Somewhere around wine 1.9.10 there was a regression in sound latency of about 20 milliseconds, using the same HelBuflen valueof 1024. This is still present in 1.9.18. I'm using pulseaudio, a 32-bit wine prefix and the game uses dsound for sound output.
However, it's very hard to notice latency for someone who isn't used to playing the game, so you'll probably have to take my word for it.
After git bisecting, I got these three patches that can't be tested due to crashes and sound glitching (probably because they're related to eachother and can't be tested individually)
There are only 'skip'ped commits left to test. The first bad commit could be any of: 569ed159b0ab31121d9aab95bf1d3ac49675f1ca 95287ac2dfb3bd64e1b65aadbdf7e375af2607d4 09890d7142a2c612b025ab9ff60358d036f9999b We cannot bisect more!
I don't know enough about the dsound internals and it doesn't seem to be as simple as reverting the patches since dsound/primary.c changed a lot since then.
I have attached logs with WINEDEBUG=+dsound of playing for a few seconds on the latest wine version (with_latency) vs commit 77184106c8c2bff7db9ea5d20b3b3330196befc1 , which should be the last one that doesn't have latency (without_latency).
https://bugs.winehq.org/show_bug.cgi?id=41329
--- Comment #1 from Franc[e]sco lolisamurai@tfwno.gf --- Created attachment 55668 --> https://bugs.winehq.org/attachment.cgi?id=55668 first 1mb of log from playing for a few seconds without latency
https://bugs.winehq.org/show_bug.cgi?id=41329
--- Comment #2 from Franc[e]sco lolisamurai@tfwno.gf --- Created attachment 55669 --> https://bugs.winehq.org/attachment.cgi?id=55669 first 1mb of log from playing for a few secs with latency on the latest wine version
https://bugs.winehq.org/show_bug.cgi?id=41329
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #3 from winetest@luukku.com --- Bug 42074 also same game.
"Introduced in wine 1.9.10, still present in 2.0-rc3. Copying my description from a comment on another issue."
You also point into 1.9.10. Also you mention the same commit
569ed159b0ab31121d9aab95bf1d3ac49675f1ca which caused that regression. It's said to be fixed 0190c50a8ca14e21355befc212b60f6409594aec which is in wine 2.0-rc4.
This is very likely already fixed. Could you test against wine 2.0rc4?
https://bugs.winehq.org/show_bug.cgi?id=41329
--- Comment #4 from Franc[e]sco lolisamurai@tfwno.gf --- Sorry, haven't had time to accurately test this until now.
I'm pretty sure the problem is still present, and as this bug report points out it appears to only affect winepulse.drv: https://bugs.winehq.org/show_bug.cgi?id=42856
we can probably continue this in that bug report.
https://bugs.winehq.org/show_bug.cgi?id=41329
--- Comment #5 from winetest@luukku.com --- (In reply to Franc[e]sco from comment #4)
Sorry, haven't had time to accurately test this until now.
I'm pretty sure the problem is still present, and as this bug report points out it appears to only affect winepulse.drv: https://bugs.winehq.org/show_bug.cgi?id=42856
we can probably continue this in that bug report.
What's your wine version currently? Latest released is 2.8.
https://bugs.winehq.org/show_bug.cgi?id=41329
--- Comment #6 from Franc[e]sco lolisamurai@tfwno.gf --- I'm using 2.8, still experiencing the same latency when using winepulse.
as reported in the other bug report, switching to winealsa (which actually still outputs to pulse via the alsa plugin) removes all the latency.
-
wine-bugs＠winehq.org