http://bugs.winehq.org/show_bug.cgi?id=30986
Bug #: 30986 Summary: Severe frame rate regression in 3D games when changing cursor graphic (esp with animated cursors) Product: Wine Version: 1.5.7 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d AssignedTo: wine-bugs@winehq.org ReportedBy: jimportal@gmail.com Classification: Unclassified
While playing Diablo 3 (D3) frame rates decrease significantly when the game uses an animated cursor. In World of Warcraft (WoW) you can see a brief hiccup in animation when you mouse over something that changes the cursor shape. It's not as severe in WoW but WoW does not use animated cursors like D3 does.
A git bisect led to:
c2c0743ccfd4a945df3b88231d62731e14da3324 is the first bad commit commit c2c0743ccfd4a945df3b88231d62731e14da3324 Author: Henri Verbeet hverbeet@codeweavers.com Date: Fri Jun 22 15:57:29 2012 +0200
wined3d: Properly retrieve the display mode in wined3d_device_set_cursor_properties().
:040000 040000 5e0f894018153ba040091ace29474ccd9cfd0288 705cc9dc4e980c64ae3edfda864a3c47a6a9beb4 M dlls
http://bugs.winehq.org/show_bug.cgi?id=30986
James Eder jimportal@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |c2c0743ccfd4a945df3b88231d6 | |2731e14da3324
http://bugs.winehq.org/show_bug.cgi?id=30986
James Eder jimportal@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |hverbeet@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=30986
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #1 from James Eder jimportal@gmail.com 2012-06-25 13:09:51 CDT --- This looks similar to Bug 30538 in that the cause seems to be that wined3d_swapchain_get_display_mode is just too slow.
The patch suggested there (also on wine patches http://source.winehq.org/patches/data/87522) avoids using wined3d_swapchain_get_display_mode. I don't know if it would be better to improve wined3d_swapchain_get_display_mode or to avoid it where speed is an issue.
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #2 from James Eder jimportal@gmail.com 2012-06-25 13:13:54 CDT --- Sorry I meant to say wined3d_get_adapter_display_mode instead of wined3d_swapchain_get_display_mode (which calls wined3d_get_adapter_display_mode).
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #3 from James Eder jimportal@gmail.com 2012-06-25 16:03:06 CDT --- Created attachment 40716 --> http://bugs.winehq.org/attachment.cgi?id=40716 Don't care about screen size in wined3d_device_set_cursor_properties().
I'm not sure if applications depend on this function failing if the cursor is bigger than the screen but removing the check altogether fixes the problem for me. I don't know if something like this could be accepted into Wine, however.
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #4 from Henri Verbeet hverbeet@gmail.com 2012-06-28 16:50:26 CDT --- Mostly for reference, on what hardware is this?
http://bugs.winehq.org/show_bug.cgi?id=30986
Mike Cronce floppie@quadra-tec.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |floppie@quadra-tec.net
--- Comment #5 from Mike Cronce floppie@quadra-tec.net 2012-06-28 22:34:36 CDT --- I can't speak for anybody else, but I'm seeing it on an GeForce GTX470 and E8500 CPU.
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #6 from James Eder jimportal@gmail.com 2012-06-28 22:52:42 CDT --- (In reply to comment #4)
Mostly for reference, on what hardware is this?
GeForce GTS 250 on NVIDIA's 302.17 drivers AMD Phenom(tm) II X4 955 Processor 8GB memory
For what it's worth the most recent NVIDIA binary drivers now support randr 1.3, so fixing bug 30184 might be the thing to do?
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #7 from Henri Verbeet hverbeet@gmail.com 2012-06-29 01:03:19 CDT --- Created attachment 40780 --> http://bugs.winehq.org/attachment.cgi?id=40780 patch
Possibly, although my impression was that nvidia wasn't affected by that. I suppose it could be a new "feature" in the new drivers that support RandR 1.3. Does the attached patch make it any better?
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #8 from James Eder jimportal@gmail.com 2012-06-29 09:57:15 CDT --- (In reply to comment #7)
Does the attached patch make it any better?
Yes, with that patch I do not notice any slowdowns.
http://bugs.winehq.org/show_bug.cgi?id=30986
Anton Romanov theli.ua@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |theli.ua@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #9 from Michael Cronenworth mike@cchtml.com 2012-07-06 23:18:35 CDT --- Created attachment 40901 --> http://bugs.winehq.org/attachment.cgi?id=40901 cursor fix for wine 1.5.8
I can also confirm the patch from comment 7 fixes the performance regression. I have attached an updated patch that applies against Wine 1.5.8.
http://bugs.winehq.org/show_bug.cgi?id=30986
Michael Cronenworth mike@cchtml.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mike@cchtml.com
http://bugs.winehq.org/show_bug.cgi?id=30986
Karl-Johan Karlsson creideiki+winebugs@lysator.liu.se changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |creideiki+winebugs@lysator. | |liu.se
--- Comment #10 from Karl-Johan Karlsson creideiki+winebugs@lysator.liu.se 2012-07-07 16:42:40 CDT --- This bug affects Fate of the World as well, where it is fixed by the patch in comment 9.
http://bugs.winehq.org/show_bug.cgi?id=30986
Jerome Leclanche adys.wh@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW CC| |adys.wh@gmail.com Ever Confirmed|0 |1
--- Comment #11 from Jerome Leclanche adys.wh@gmail.com 2012-07-11 19:14:48 CDT --- Confirming presence in wine-1.5.8-100-g4a01368. Patch no longer applies.
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #12 from msthev ms.pub@gmx.com 2012-07-12 01:56:39 CDT --- Starcraft II is also affected: bug 31190 and bug 31150
As for video card, I'm using NVidia GF 8800GT with proprietary drivers. At the same time it seems that the author of bug 31190 is on ATI.
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #13 from Karl-Johan Karlsson creideiki+winebugs@lysator.liu.se 2012-07-12 07:37:27 CDT --- For completeness: my hardware for the result in comment 10 is an Intel Sandybridge Mobile (GT2+) GPU in a Core i7 2640M CPU.
http://bugs.winehq.org/show_bug.cgi?id=30986
Jerome Leclanche adys.wh@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |petkra@chem.gla.ac.uk
--- Comment #14 from Jerome Leclanche adys.wh@gmail.com 2012-07-12 09:49:55 CDT --- *** Bug 31190 has been marked as a duplicate of this bug. ***
http://bugs.winehq.org/show_bug.cgi?id=30986
Jerome Leclanche adys.wh@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |binbiniqegabenik@gmail.com
--- Comment #15 from Jerome Leclanche adys.wh@gmail.com 2012-07-12 09:50:04 CDT --- *** Bug 31150 has been marked as a duplicate of this bug. ***
http://bugs.winehq.org/show_bug.cgi?id=30986
Sven sven.wine@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sven.wine@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=30986
James Eder jimportal@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |2b4a7912ec00e1c69e98ccb1d2e | |810d8df572027 Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #16 from James Eder jimportal@gmail.com 2012-07-16 15:32:52 CDT --- Fixed in today's git.
http://bugs.winehq.org/show_bug.cgi?id=30986
--- Comment #17 from James Eder jimportal@gmail.com 2012-07-16 16:16:56 CDT --- For the sake of accounting this could be marked as a duplicate of bug 30184 or perhaps a dependency on that bug.
I found bug 30184 only after filing this one. I left his bug open because at the time I wasn't sure if the problem would be resolved by implementing randr 1.3 support or by some other solution.
I suspect those that maintain the Wine Bugzilla will know better than I (if such things matter).
http://bugs.winehq.org/show_bug.cgi?id=30986
Jerome Leclanche adys.wh@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- See Also| |http://bugs.winehq.org/show | |_bug.cgi?id=30184
http://bugs.winehq.org/show_bug.cgi?id=30986
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #18 from Alexandre Julliard julliard@winehq.org 2012-07-17 13:52:32 CDT --- Closing bugs fixed in 1.5.9.