http://bugs.winehq.org/show_bug.cgi?id=9768
Summary: Performance Regression in 3Dmark03 Product: Wine Version: CVS/GIT Platform: PC-x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: wine-directx-d3d AssignedTo: wine-bugs@winehq.org ReportedBy: greg87@online.de
In current wine there is a loss of about 1500points.
Looking at the detailed results, most tests are on the same speed. Except Trolls Lair and Battle for Proxycon. And in the additional tests the Ragtroll bench.
Caused by: commit ad3fe8562b1b4e71b18d0c69a77caad8189c3d28 Author: Stefan Dösinger stefan@codeweavers.com Date: Fri Sep 14 12:00:46 2007 +0200
wined3d: Enable the stencil test if two sided stencil is used.
I verified the correctness of the regression test by reverting it in the source code.
Im running on nvidia drivers 100.14.19 and a Geforce 8600GT
http://bugs.winehq.org/show_bug.cgi?id=9768
Gregor Münch greg87@online.de changed:
What |Removed |Added ---------------------------------------------------------------------------- URL| |http://www.futuremark.com/do | |wnload/3dmark03/ Keywords| |download, regression
http://bugs.winehq.org/show_bug.cgi?id=9768
Gregor Münch greg87@online.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |stefandoesinger@gmx.at
--- Comment #1 from Gregor Münch greg87@online.de 2007-09-25 18:47:51 --- adding author of the patch
http://bugs.winehq.org/show_bug.cgi?id=9768
--- Comment #2 from Stefan Dösinger stefandoesinger@gmx.at 2007-09-26 05:14:40 --- I am tempted to blame the driver for this. 1500 3Dmarks loss is a very massive loss, I don't think plainly enabling the stencil test can cause this. It may force the driver to fall back to software rendering for some reason. The Geforce 8 drivers are known to have big performance issues on Linux. Do you have a geforce 7 or geforce 6 to test with?
What is interesting too however, is that the stencil test was incorrectly disabled before, and the app still worked. This may be an attacking point for a workaround of the driver bug(if it is one).
http://bugs.winehq.org/show_bug.cgi?id=9768
--- Comment #3 from Stefan Dösinger stefandoesinger@gmx.at 2007-10-12 02:05:03 --- Could it be that the shadows in the game benchmarks were fixed with this patch?
Can you retest this bug? Henri ran 3DMark2003 on Wineconf, and it was as fast as ever before.
http://bugs.winehq.org/show_bug.cgi?id=9768
--- Comment #4 from Stefan Dösinger stefandoesinger@gmx.at 2007-10-24 05:50:24 --- Henri said that the shadows started working at some point. Could it be that this patch fixed the shadows?
http://bugs.winehq.org/show_bug.cgi?id=9768
Gregor Münch greg87@online.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |INVALID
--- Comment #5 from Gregor Münch greg87@online.de 2007-11-12 12:16:32 --- I think I didn saw any optical differences but the fact that my 7600GT also looses some points (from 9900 to 9300) leads me to the conclusion that there is no wine bug.
http://bugs.winehq.org/show_bug.cgi?id=9768
Dan Kegel dank@kegel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #6 from Dan Kegel dank@kegel.com 2008-01-28 05:58:45 --- Closing all RESOLVED INVALID older than four weeks.
http://bugs.winehq.org/show_bug.cgi?id=9768
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Version|CVS/GIT |unspecified
-
wine-bugs＠winehq.org