http://bugs.winehq.org/show_bug.cgi?id=29079
Bug #: 29079 Summary: Serious Sam HD TSE Performance Regression Product: Wine Version: 1.3.32 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: FinalCrisisSX@Hotmail.com Classification: Unclassified
Arch Linux x86_64, up to date except for Xorg (using 1.10.4) and Nvidia (using 280.13) packages due to a transparency bug in TF2 I've reported to nvidia caused by nvidia drivers 285.05. In Serious Sam HD: The Second Encounter, at some point between 1.3.18 and 1.3.32, the game suffers a huge performance drop. It's odd in that if I play the game after the bad commit, it seems to have small bursts of working fine and running smoothly then starts lagging out / chugging along again shortly. With 1.3.18 however, it works fine continuously. I ran a regression test and found the culprit. The bad commit description suggests drivers over 270.xx should be fine, but it isn't the case for this particular game for me. Other games suffer no such problems. Indeed, if I use the patch remove-GL_ARB_map_buffer_range.patch from http://bugs.winehq.org/show_bug.cgi?id=27600 this bug thread, the game no longer has the performance issue, as one might expect.
68b15bc5ffe6ddf5d08cbc13479eaf718ad5e39f is the first bad commit commit 68b15bc5ffe6ddf5d08cbc13479eaf718ad5e39f Author: Stefan Dösinger stefan@codeweavers.com Date: Tue Apr 19 21:24:26 2011 +0200
wined3d: Give GL_ARB_map_buffer_range another try.
Nvidia's 270.xx.yy driver series fix the glMapBuffer alignment issues that caused major performance problems. If the user is using an older driver we now drop the VBO instead of using doublebuffered loading, which means that we fall back to the current behavior(no dynamic VBO) as needed. Dynamic VBOs are needed on Nvidia drivers for GL_ARB_instanced_arrays.
:040000 040000 0cada15043b731febf2ef2a673aaaecc4af75f8d c474ade74a8db2a9c547674dbb4c809841087a3d M dlls
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #1 from Brandon Howard FinalCrisisSX@Hotmail.com 2011-11-13 01:16:36 CST --- Also, if it is of any importance, my nvidia card is a GTS 250.
http://bugs.winehq.org/show_bug.cgi?id=29079
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression Regression SHA1| |68b15bc5ffe6ddf5d08cbc13479 | |eaf718ad5e39f
http://bugs.winehq.org/show_bug.cgi?id=29079
Stefan Dösinger stefan@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |stefan@codeweavers.com
--- Comment #2 from Stefan Dösinger stefan@codeweavers.com 2011-11-13 14:15:17 CST --- Are you using multiple monitors? The Nvidia driver has severe problems with GL_ARB_map_buffer_range and multiple monitors that are reproducible in Linux apps(Bug 27959).
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #3 from Brandon Howard FinalCrisisSX@Hotmail.com 2011-11-13 16:56:02 CST --- (In reply to comment #2)
Are you using multiple monitors? The Nvidia driver has severe problems with GL_ARB_map_buffer_range and multiple monitors that are reproducible in Linux apps(Bug 27959).
Nope, just one monitor.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #4 from Brandon Howard FinalCrisisSX@Hotmail.com 2011-11-18 16:57:59 CST --- Sort of good news! I tested Serious Sam HD: The First Encounter, and the lag shows up there too. It is more constant than in TSE for whatever reason.
The good news is, a playable demo is here that would probably have the same issue as the retail: http://store.steampowered.com/app/41000/
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #5 from Brandon Howard FinalCrisisSX@Hotmail.com 2011-12-03 14:11:29 CST --- Created attachment 37773 --> http://bugs.winehq.org/attachment.cgi?id=37773 Patch to revert
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #6 from Brandon Howard FinalCrisisSX@Hotmail.com 2011-12-03 14:20:53 CST --- For those who were using the previously mentioned patch to fix samhd, it no longer patches correctly since some point between 1.3.33 and 1.3.34. Using the actual bad commit itself however as a patch with 'patch -p1 -R' (revert) works fine and the game runs smoothly again minus the issues from bug #29080. The commit patch is attached up at the top of this page for your convenience, but I screwed up when posting it and forgot the 'patch' check box. It doesn't seem to matter though, outside of clicking the link above this particular post not showing up as I expected it to..
http://bugs.winehq.org/show_bug.cgi?id=29079
Dmitry Timoshkov dmitry@baikal.ru changed:
What |Removed |Added ---------------------------------------------------------------------------- Summary|Serious Sam HD TSE |Serious Sam HD TSE: low |Performance Regression |performance Severity|normal |minor
http://bugs.winehq.org/show_bug.cgi?id=29079
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |directx-d3d
http://bugs.winehq.org/show_bug.cgi?id=29079
Gediminas Jakutis vinislentoje@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |vinislentoje@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=29079
RealGecko alexandrbezenkov@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |alexandrbezenkov@gmail.com
--- Comment #7 from RealGecko alexandrbezenkov@gmail.com 2012-01-28 08:58:24 CST --- I have Arch Linux and Nvidia GT 430 with Nvidia 290.10 driver and can confirm this bug under wine 1.3.37. Even setting game settings to low doesn`t solve performance problem turning game almost unplayable. However if I run the game on second X server (through xgame script) performance problem is gone and game runs smooth and nice.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #8 from RealGecko alexandrbezenkov@gmail.com 2012-01-28 09:04:39 CST --- (In reply to comment #7)
I have Arch Linux and Nvidia GT 430 with Nvidia 290.10 driver and can confirm this bug under wine 1.3.37. Even setting game settings to low doesn`t solve performance problem turning game almost unplayable. However if I run the game on second X server (through xgame script) performance problem is gone and game runs smooth and nice.
But not always....
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #9 from Stefan Dösinger stefan@codeweavers.com 2012-02-27 08:20:52 CST --- Can you try attachment 39097 [details](from bug 30019)? I tested it with the demo, but could not get conclusive results - the game keeps changing the settings, which makes testing difficult.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #10 from Brandon Howard FinalCrisisSX@Hotmail.com 2012-03-02 10:51:58 CST --- (In reply to comment #9)
Can you try attachment 39097 [details](from bug 30019)? I tested it with the demo, but could not get conclusive results - the game keeps changing the settings, which makes testing difficult.
Quick testing and a once-over to make sure I didn't do anything stupid in testing the patch finds that it does indeed fix the problem. SamHD TFE and SamHD TSE both seem to run smoothly. Wine 1.4-rc5 = lag, wine 1.4-rc5 + your patch = no lag. \o/
Initially, I couldn't get the games to run again.. apparently they don't like quartz forced via winetricks. They worked after undoing that. Neither would run in a fresh prefix any more for some reason >_<. Either way, that's unrelated to this bug report, but just in case..
RealGecko, care to add a second confirmation?
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #11 from Gediminas Jakutis vinislentoje@gmail.com 2012-06-25 04:34:44 CDT --- This bug is no longer present after 1.5.x (can't remember on which version exactly). On my machine, at least.
http://bugs.winehq.org/show_bug.cgi?id=29079
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED CC| |wylda@volny.cz Resolution| |FIXED
--- Comment #12 from Wylda wylda@volny.cz 2012-06-27 00:16:52 CDT ---
Reported fixed.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #13 from Brandon Howard FinalCrisisSX@Hotmail.com 2012-06-27 18:07:28 CDT --- This is not fixed for me in 1.5.7. Same trouble as I reported in the OP without a patch / revert.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #14 from Gediminas Jakutis vinislentoje@gmail.com 2012-06-28 05:49:42 CDT --- (In reply to comment #13)
It seem You are right. Sorry, my bad.
For some reason, I noticed that the game starts in three possible ways: 1 - Runs in full performance. 2 - Runs at approx. half the "full" performance with short severe framerate drop when changing weapons. 3 - Exactly as the bug description (constantly switches between running full speed and slowing down to a crawl)
I just probably been /really/ lucky with constantly getting the 1st one for a whole week; giving me a false impression of the bug no longer being present.
http://bugs.winehq.org/show_bug.cgi?id=29079
Brandon Howard FinalCrisisSX@Hotmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |UNCONFIRMED Resolution|FIXED |
--- Comment #15 from Brandon Howard FinalCrisisSX@Hotmail.com 2012-06-28 13:38:40 CDT --- (In reply to comment #14)
(In reply to comment #13)
Yeah, I've had occasional weird success too.
I don't know if I can unmark this fixed or if an admin must do it.. let's see if this does the trick.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #16 from Brandon Howard FinalCrisisSX@Hotmail.com 2012-10-13 04:08:39 CDT --- Just an update since this report hasn't been commented on for a while: this bug does still apply to wine 1.5.15.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #17 from Gediminas Jakutis gediminas@varciai.lt 2013-02-04 09:24:12 CST --- Played a lot of this game today and did not notice any signs of this bug at all. On Wine-1.5.23 and nvidia drivers v313.09 Others, could You check too? If it's not only for me, then we can probably close this one.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #18 from RealGecko alexandrbezenkov@gmail.com 2013-02-05 08:02:32 CST --- (In reply to comment #17)
Played a lot of this game today and did not notice any signs of this bug at all. On Wine-1.5.23 and nvidia drivers v313.09 Others, could You check too? If it's not only for me, then we can probably close this one.
This happens sometimes, you can play game smoothly and nice today, but you won`t be able to play at all tomorrow. The nature of this bug is mysterious :D
http://bugs.winehq.org/show_bug.cgi?id=29079
Stefan Dösinger stefan@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #19 from Stefan Dösinger stefan@codeweavers.com 2013-02-05 08:12:09 CST --- I expect this patch to be fixed by 1835e2f5baa376eab95c760951fcc6a3612aa6cd, on cards that support ARB_instanced_arrays. There were similar bugs that I expected to be fixed by this patch, and they weren't, so I kept all of them open for now.
I'll leave this bug open for a while and close it if nobody can confirm that it still occurs.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #20 from Frozen Fox FinalCrisisSX@Hotmail.com 2013-02-26 15:01:03 CST --- (In reply to comment #19)
I expect this patch to be fixed by 1835e2f5baa376eab95c760951fcc6a3612aa6cd, on cards that support ARB_instanced_arrays. There were similar bugs that I expected to be fixed by this patch, and they weren't, so I kept all of them open for now.
I'll leave this bug open for a while and close it if nobody can confirm that it still occurs.
Unfortunately, I have rebuilt my machine since the original post and accordingly cannot be sure it works as expected under those specs. I don't see the same trouble on the new machine though. The performance isn't nearly as high as it should be with my significantly stronger machine, but I don't see the same dramatic fluctuation in FPS as reported. So, fixed? Maybe? RealGecko, can you confirm?
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #21 from Gediminas Jakutis gediminas@varciai.lt 2013-02-27 04:22:14 CST --- (In reply to comment #20) Before You start suspecting this bug is back: how's Your CPU configured?
SSHD does not do well with power saving features and is highly CPU bottlenecked. Unless I force my CPU to the highest power state, the performance is underwhelming. Meanwhile, if I overclock my CPU enough, it runs as fast as ~75% native's performance, where it starts getting GPU bound.
P.S. This engine does not do well with Linux's power saving && core switching logic when run natively [as SS3] and that makes the game run much slower. The engine parts that suffer from this are unchanged between SSHD and SS3. Thus, it's logical it suffers from this even when there's Wine involved. Source: a chat with a dev from Croteam.
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #22 from RealGecko alexandrbezenkov@gmail.com 2013-02-27 07:52:13 CST --- Unfortunately I moved to another city and left my PC behind. So I cannot test anymore :(
http://bugs.winehq.org/show_bug.cgi?id=29079
--- Comment #23 from Gediminas Jakutis gediminas@varciai.lt 2013-03-16 11:50:59 CDT --- (In reply to comment #18)
Played it for over 50 hours in the past four weeks and did not encounter this bug at all, so I'd say it's fixed. Thus, can we close this now?
(always play on the latest git, drivers: 313.18, Card: GF GTX 560)
http://bugs.winehq.org/show_bug.cgi?id=29079
Stefan Dösinger stefan@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #24 from Stefan Dösinger stefan@codeweavers.com 2013-03-17 11:25:00 CDT --- reported fixed
http://bugs.winehq.org/show_bug.cgi?id=29079
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #25 from Alexandre Julliard julliard@winehq.org 2013-03-29 13:33:09 CDT --- Closing bugs fixed in 1.5.27.
-
wine-bugs＠winehq.org