2008/10/9 Zachary Goldberg zgold@bluesata.com:
And yes I do think this data would be valuable. Especially for the d3d people who have a very rough idea of what to focus on (from what they've told us at wineconf).
That's not quite what Stefan said, actually. The issue is that while there are obviously some games that are very popular like World of Warcraft, Call of Duty, Counter-Strike, etc., there simply isn't a single application that almost everyone runs. Another issue is that compared to regular applications, games have a relatively short lifespan. If a game remains popular for an entire year that's a relatively long time already. So it's not so much that we don't know what to focus on, but rather that it makes more sense to work on things like "fix surface management" or "implement dualhead" than on "make Red Alert 3 work perfectly", although those aren't mutually exclusive, of course. You also have to consider that debugging a problem often takes just as much or more time as implementing a fix, which is a bit of a waste if you already know there are parts of the code that have problems which will probably cause a bug in some application.