http://bugs.winehq.org/show_bug.cgi?id=20498
Ville Aakko ville.aakko@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ville.aakko@gmail.com
--- Comment #8 from Ville Aakko ville.aakko@gmail.com 2010-06-05 09:24:16 --- Hi,
I can confirm this bus since I've been having the same problem for some time. I finally got the time to make the regression testing via git-bisect (it didn't take that long after all). The result: " #git bisect bad b5da7f49d332ed5d64be1256e14035af74a7efa2 is the first bad commit commit b5da7f49d332ed5d64be1256e14035af74a7efa2 Author: Henri Verbeet hverbeet@codeweavers.com Date: Thu Jul 2 10:01:37 2009 +0200
wined3d: Don't reuse random GL contexts during initialization.
Obviously there's no guarantee about the state of such a context. The specific problem is that it might have GL_UNPACK_CLIENT_STORAGE_APPLE enabled, causing some glTexImage2D() calls to fail, but it's a bad idea in general.
:040000 040000 91e52fba45ce42a6e4c1fce4281308745aea44cf ae1eecf091bdc72d34e708399ebd1670b198e00f M dlls "
I only tested the game main menu after every compile. If you need the git bisect log I will provide it.
FWIW, even with a recent wine (with the regression) I noticed that if I run it in a window (instead of full screen - i.e. either a virtual dekstop smaller than the real resolution, or set Civilization IV to run in a window), the rendering isn't corrupted as in the screenshot. Instead, some 3D models (most notably fishing boats) are corrupted and cause weird artifacts on the screen. But even then the game is playable.
Hope this helps!