http://bugs.winehq.org/show_bug.cgi?id=8615
--- Comment #8 from Rodney Moss edge@rsm.id.au 2007-10-08 04:04:26 --- Created an attachment (id=8460) --> (http://bugs.winehq.org/attachment.cgi?id=8460) Console log for wine 0.9.35
Further testing:
wine 0.9.34, clean install, seems to have the same problem as 0.9.46
wine 0.9.35 seems to crash at the same point as 0.9.46, for me, with slight differences. The graphical glitches on the campaign loading screen clearly show possible attempts at multiple windows open, instead of just blackness. This might be be bug 2398 but i don't know enough to say. The console log reports (mis)handling the back buffering slightly differently.
wine 0.9.28 'worked' once of about 10 starts, crashing shortly into the turn rather than immediately. However, 0.9.46 (seemingly) randomly worked for me a few times too, too, and worked better.
Regardless, i can't regression test since i can't find a version with the campaign working any better than 0.9.46. It may be the versions that used to 'work' did so with different nvidia drivers, and/or different X versions?