http://bugs.winehq.org/show_bug.cgi?id=10414
Summary: Minimap in Warhammer 40k: Dark Crusade is offset Product: Wine Version: 0.9.49. Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: enhancement Priority: P2 Component: wine-x11driver AssignedTo: wine-bugs@winehq.org ReportedBy: fdelente@mail.cpod.fr
Created an attachment (id=9106) --> (http://bugs.winehq.org/attachment.cgi?id=9106) Part of screenshot showing the bug
In this game, the minimap used to work ok. Since three or four wine versions, the area that should be lit stays gray; the lit portion is offset to the right.
For example, if my troop are in the left corner of the minimap, the lit area is in the center; if my troops are in the center, the lit area is on the right. If they are in the center, nothing is lit.
The offset is only horizontal, the vertical position of the lit area is correct.
On the attached PNG, the bug is clearly shown: allied troops in the left area should be surrounded by the lit area that is in the center.
http://bugs.winehq.org/show_bug.cgi?id=10414
Alexander Nicolaysen Sørnes alex@thehandofagony.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |alex@thehandofagony.com URL| |http://www.3dgamers.com/dlse | |lect/games/wh40kdawnwardc/da | |rkcrusadedemo.exe.html Status|UNCONFIRMED |NEW Component|wine-x11driver |wine-directx-d3d Ever Confirmed|0 |1 Keywords| |download, regression
--- Comment #1 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2007-11-11 16:15:48 --- Confirming with the Dark Crusade demo that this is a problem in Wine 0.9.48. Please do regression testing as described here
http://wiki.winehq.org/GitWine#head-fc6d68a85d32bb6a8f4ea0485f50dbb6409ab6f4
To find out which change caused the regression.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #2 from F. Delente fdelente@mail.cpod.fr 2007-11-12 15:13:43 --- (In reply to comment #1)
Confirming with the Dark Crusade demo that this is a problem in Wine 0.9.48. Please do regression testing as described here
http://wiki.winehq.org/GitWine#head-fc6d68a85d32bb6a8f4ea0485f50dbb6409ab6f4
To find out which change caused the regression.
I started to do the regression but had to stop because of an error in compilation after a few bisects:
directx.c: In function 'IWineD3DImpl_FillGLCaps': directx.c:820: error: 'CARD_NVIDIA_GEFORCE_8800GTS' undeclared (first use in this function) directx.c:820: error: (Each undeclared identifier is reported only once directx.c:820: error: for each function it appears in.) directx.c:823: error: 'CARD_NVIDIA_GEFORCE_8600GT' undeclared (first use in this function) directx.c:827: error: 'CARD_NVIDIA_GEFORCE_8300GS' undeclared (first use in this function) make[2]: *** [directx.o] Error 1 make[2]: Leaving directory `/usr/local/testSpace/wine/wine-git/wine-git/dlls/wined3d' make[1]: *** [wined3d/__install-lib__] Error 2
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #3 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2007-11-12 15:22:57 --- Try moving a few commits away (explained in the guide)
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #4 from F. Delente fdelente@mail.cpod.fr 2007-11-12 15:39:23 --- (In reply to comment #3)
Try moving a few commits away (explained in the guide)
Sorry but I didn't see anything relevant.
Do I have to do something like
git reset --hard HEAD~5 # go back 5 commits ?
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #5 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2007-11-12 16:58:36 --- Yes
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #6 from F. Delente fdelente@mail.cpod.fr 2007-11-13 10:57:40 --- (In reply to comment #5)
Yes
I started with
git bisect good wine-0.9.44 git bisect bad wine-0.9.45
and after a few bisects I got
fe0f0eb48a12e29af6a9e7407d4eec8bc500a057 is first bad commit commit fe0f0eb48a12e29af6a9e7407d4eec8bc500a057 Author: Stefan Dösinger stefan@codeweavers.com Date: Mon Aug 27 14:33:46 2007 +0200
wined3d: Some texture transform fixes.
:040000 040000 ff0c54acd4fd254bae48ade26e8ebd095bbd8af9 4675067c38ea90f92f13f5c8c92d44439d3078ea M dlls
The version of wine that I compiled last is
wine-0.9.44-ga8e21d0
I couldn't add stefan@codeweavers.com in the CC, this address seems to be invalid(?)
http://bugs.winehq.org/show_bug.cgi?id=10414
Alexander Nicolaysen Sørnes alex@thehandofagony.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |stefandoesinger@gmx.at
--- Comment #7 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2007-11-13 11:32:57 --- He sometimes uses a different address. Thanks for the regression test!
Adding patch author to CC list.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #8 from Stefan Dösinger stefandoesinger@gmx.at 2007-11-13 12:33:07 --- I subscribed to bugzilla with my @gmx address long before I joined codeweavers, but I am sending my patches from the cw address because the @gmx address is sometimes slow
So yet another texture transform regression... I have another one on my list. Can you keep an eye on bug 9797? A fix for that bug might fix this bug here too, but it is not necessarily a duplicate.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #9 from F. Delente fdelente@mail.cpod.fr 2007-11-16 10:23:08 --- Since bug #9797 was marked as solved today, I tried today's git (wine tag wine-0.9.49-ga529ef4) but the minimap bug is still here.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #10 from F. Delente fdelente@mail.cpod.fr 2007-12-22 14:39:57 --- Bug still present in wine 0.9.51, git version (tag wine-0.9.51-g16aadb2).
http://bugs.winehq.org/show_bug.cgi?id=10414
escapedturkey escapedturkey@escapedturkey.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |escapedturkey@escapedturkey. | |com
--- Comment #11 from escapedturkey escapedturkey@escapedturkey.com 2008-01-17 21:26:30 --- This is also a problem with the SoulStorm demo. Any chance this can be fixed?
The latest version of Wine (0.9.53) runs DC and SoulStorm perfectly (loading times are super fast). It's just this minimap bug that still exists.
http://bugs.winehq.org/show_bug.cgi?id=10414
Danila Sentiabov dsent@mail.ru changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |dsent@mail.ru
--- Comment #12 from Danila Sentiabov dsent@mail.ru 2008-01-24 06:50:06 --- Confirming this bug in 0.9.53. Very annoying.
http://bugs.winehq.org/show_bug.cgi?id=10414
Phil pme991@cox.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |pme991@cox.net
--- Comment #13 from Phil pme991@cox.net 2008-02-07 23:32:38 --- Also Confirming this bug in 0.9.54. Used Dawn of War.
http://bugs.winehq.org/show_bug.cgi?id=10414
Marcin ergo14@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ergo14@gmail.com
--- Comment #14 from Marcin ergo14@gmail.com 2008-03-17 04:32:39 --- I can confirm this in wine 0.9.56 and 0.9.57
http://bugs.winehq.org/show_bug.cgi?id=10414
Erik Andresen erik@vontaene.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |erik@vontaene.de
--- Comment #15 from Erik Andresen erik@vontaene.de 2008-03-18 13:06:48 --- ok, took me a few hours, but:
I also did some regression testing. I also came to the conclusion, that this patch broke the Minimap: http://source.winehq.org/git/wine.git/?a=commitdiff_plain;h=fe0f0eb48a12e29a...
Now, please note that I have no idea, what the float named mat is, or what I was doing, but by doing some testing and closely looking at the patch I found a fix that is working for me.
With this patch with applies against Wine 0.9.57, the minimap completely works against for me:
--- a/dlls/wined3d/utils.c +++ b/dlls/wined3d/utils.c @@ -2609,6 +2609,8 @@ void set_texture_matrix(const float *smat, DWORD flags, BO * 1.0 doesn't do anything bad. The caller will set this value */ case WINED3DDECLTYPE_UNUSED: /* No texture coords, 0/0/0/1 defa + mat[12] = mat[8]; + mat[13] = mat[9]; break; default: FIXME("Unexpected fixed function texture coord input\n");
Would be good if someone can confirm this one.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #16 from Stefan Dösinger stefandoesinger@gmx.at 2008-03-18 13:08:40 --- I think this needs a test. It looks kinda odd, but I cannot prove it wrong either. See dlls/d3d9/tests/visual.c, especially the texture_transform_flags test.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #17 from Erik Andresen erik@vontaene.de 2008-03-18 13:16:55 --- Created an attachment (id=11477) --> (http://bugs.winehq.org/attachment.cgi?id=11477) possible fix
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #18 from Erik Andresen erik@vontaene.de 2008-03-18 13:17:58 --- Sorry, Had a problem with pasting the patch, thats why a few characters are missing. I attached the file to this bug instead.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #19 from F. Delente fdelente@mail.cpod.fr 2008-03-18 13:37:01 --- (In reply to comment #18)
Sorry, Had a problem with pasting the patch, thats why a few characters are missing. I attached the file to this bug instead.
I tried the patch, actually it seems to solve the minimap problem; however I noticed that the 'Start mission' text (just when beginning a Skirmish map) is missing on the button.
Great hint anyway :-)
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #20 from Erik Andresen erik@vontaene.de 2008-03-19 03:20:41 --- Strange. My Start Mission -text is still there.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #21 from F. Delente fdelente@mail.cpod.fr 2008-03-19 09:45:03 --- (In reply to comment #20)
It happened only once, the text appears correctly now.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #22 from SF simonfitz@yahoo.com 2008-03-23 17:12:43 --- (In reply to comment #17)
Created an attachment (id=11477)
--> (http://bugs.winehq.org/attachment.cgi?id=11477) [details]
possible fix
I can confirm that this patch does make the minimap work properly - tested using v0.9.58 running on Ubuntu 7.04 with DoW Soulstorm. I didn't notice any missing text on buttons.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #23 from escapedturkey escapedturkey@escapedturkey.com 2008-03-23 18:09:53 --- Will this patch be included in the next version of Wine? :)
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #24 from Stefan Dösinger stefandoesinger@gmx.at 2008-03-23 19:08:30 --- No, I am opposed to including this patch without a test. Unfortunately I am a bit too busy at the moment to write one myself.
http://bugs.winehq.org/show_bug.cgi?id=10414
Georgi Chulkov godji@metapenguin.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |godji@metapenguin.org
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #25 from F. Delente fdelente@mail.cpod.fr 2008-04-08 15:27:52 --- Problem seems solved in wine 0.9.59
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #26 from Stefan Dösinger stefandoesinger@gmx.at 2008-04-08 15:30:11 --- F. Delente, do you have a radeon 8500 to 9200 card? Otherwise I can't explain how the bug could have been fixed.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #27 from F. Delente fdelente@mail.cpod.fr 2008-04-08 15:51:06 --- (In reply to comment #26)
F. Delente, do you have a radeon 8500 to 9200 card?
No, I tested on an Nvidia 7600Go.
Otherwise I can't explain how the bug could have been fixed.
Funny, since a fix was proposed a bit of time ago (even though you were opposed to iit in comment #24), I thought it had been integrated silently in 0.9.59...
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #28 from F. Delente fdelente@mail.cpod.fr 2008-04-08 15:52:11 --- My bad, I mismatched wine versions; I used 0.9.58 with the proposed fix, not a pristine 0.9.59.
Sorry!
http://bugs.winehq.org/show_bug.cgi?id=10414
Alexandre Lima formulaire@bacatta.name changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |formulaire@bacatta.name
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #29 from Alexandre Lima formulaire@bacatta.name 2008-05-13 14:15:28 --- Patch "possible fix" tested OK with wine 1.0 rc1. Congrats !
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #30 from Georgi Chulkov godji@metapenguin.org 2008-05-13 18:26:05 --- I'm sorry, I don't understand: is the fix in 1.0-rc1 already, or will it be there only in 1.0? Thanks!
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #31 from Danila Sentiabov dsent@mail.ru 2008-05-14 01:07:23 --- (In reply to comment #30)
I'm sorry, I don't understand: is the fix in 1.0-rc1 already, or will it be there only in 1.0? Thanks!
It is not in 1.0-rc1 and it will not be in 1.0. You can only apply it manually to source code and then compile wine yourself.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #32 from Stefan Dösinger stefandoesinger@gmx.at 2008-05-14 02:44:45 --- Nobody said it will not be in 1.0. However, we need a test for this patch to show that it is correct, then it can be committed. Unfortunately I am a bit out of time at the moment, so I probably won't be able to write a test.
If anyone is interested, take a look at dlls/d3d9/tests/visual.c, the texture_transform_flags test.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #33 from Erik Andresen erik@vontaene.de 2008-05-14 07:44:32 --- I will try to take a look in the next few days.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #34 from Erik Andresen erik@vontaene.de 2008-06-07 06:25:34 --- hmm. Where can I find the real documentation for the file dlls/d3d9/tests/visual.c?
e.g, when it says visual.c:2864: Test failed: quad 1 has color 00000000, expected 0x0000ff00
Code: color = getPixelColor(device, 160, 360); ok(color == 0x0000ff00, "quad 1 has color %08x, expected 0x0000ff00\n", color);
Where can I find how the value 0x0000ff00 is calculated?
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #35 from Stefan Dösinger stefandoesinger@gmx.at 2008-06-07 07:14:54 --- This file performs a bunch of Direct3D calls which result in something being rendered. The caclulation then runs through d3d9.dll, wined3d.dll, the graphics driver, graphics card and the result is read back from the graphics card.
What you probably want is the Direct3D documentation. You can find it on www.msdn.com, but the msdn docs are incomplete and sometimes wrong, and usually doesn't describe corner cases. So the more detailed documentation is the visual.c files. It essentially says DO A, DO B, DO C, RESULT is D. It doesn't say how or why because Windows is a black box. The challenge here is to find a scheme from the results(without looking at the Windows assembly code itself, since that is illegal).
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #36 from Erik Andresen erik@vontaene.de 2008-06-07 12:37:23 --- I had a feeling you say that...
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #37 from Danila Sentiabov dsent@mail.ru 2008-08-28 16:24:15 --- Bug is gone for me with wine 1.1.3.
http://bugs.winehq.org/show_bug.cgi?id=10414
--- Comment #38 from Danila Sentiabov dsent@mail.ru 2008-08-28 17:07:46 --- Checked previous versions - bug was still present in 1.1.1, but was fixed in 1.1.2.
http://bugs.winehq.org/show_bug.cgi?id=10414
Alexander Nicolaysen Sørnes alex@thehandofagony.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #39 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2008-08-28 17:14:12 --- reported fixed
http://bugs.winehq.org/show_bug.cgi?id=10414
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #40 from Alexandre Julliard julliard@winehq.org 2008-09-05 10:38:42 --- Closing bugs fixed in 1.1.4.