To view parent comment, click here.
To read all comments associated with this story, please click here.
Going after gamers is a no-go. Games are horrible applications to support, and the amount of profit to be made, for an OS vendor, in the gaming market is next to null. Who would switch to Linux for gaming, and run the constant risk of incompatibility with the latest game, just to get a few more FPS? If those few more FPS are even achievable is another question. Windows might not be a very fast OS, but its good at getting out of the way. Windows games are, in general, not OS limited, but hardware and driver limited. To make a game run faster in Linux than in Windows, you need to create hardware drivers that are better than ATI's and NVIDIA's own drivers. This is possible (IIRC: the R200 DRI drivers are now significantly faster on Quake III than fglrx on R200 hardware), but its a very difficult and thankless task, especially without hardware documentation.
The point isn't to outdo windows, the point is to offer an alternative to Windows and let Linux outdo Windows with it's other attributes i.e. better security. I know plenty of people for whom Linux offers everything they want except games, and when you are consistently multibooting into Windows for games, you might as well just install the various open source apps on Windows and go from there instead. I can't see much room for anymore Linux growth in the home market if this hurdle isn't overcome.
Problem with today graphic card is their brute force nature. What about supporting alternative such as http://www.powervr.com/







Member since:
2005-11-12
Linux has already captured it's first potential market the IT folks. Now in order to go through its next stage of growth it needs to capture the next group of potential users that is gamers i.e. the non-profesional/ultra-savy power users. To do this you need the latest games running at speeds better than or equivlent to Windows, neither of which Linux currently acheives.
Now, after considering it, it would seem to me that the best solution is to offer game developers a comprohensive framework on which they can build there game engines, and of course it should just "happen" that this framework is cross-platform compatible. It should also include an install system and a simple framework so most of their code doesn't need to be cross compiled but rather they only need an executable "stub" provided by the framework for each platform allowing them to distribute a cross platform compatible game without having to produce different versions. Oh and this framework should include comprehensive copy protection technologies, that way you have a big trump card in terms of cost and they can be assured that there game won't be pirated on non-windows platforms.