> Heh, I think you misunderstand how free drivers work. You're caught > up in the Windows model. No, I'm not. In fact, I've never used Windows at home, ever. Not once. I started using Linux in 1996; prior to that, I didn't even have a home computer, living entirely on SunOS and various SYSV machines at work. > Free software drivers work like this: > > 1) Vendor provides a sample implementation or hardware documentation > to someone or some organization > 2) That individual (who usually has nothing to do with the > manufacturer!) produces a driver > 3) You communicate to the hardware vendor and they say "WTF, we didn't > write that dude...", AND THEY ARE RIGHT Except this is wrong in the Matrox case, since the drivers were written by Matrox. This is why Matrox got so popular in the first place -- they *wrote free drivers for Linux*. But along the way, after several years of maintaining them, they decided to no longer do so. In the meantime, the DRI project's one expert on Matrox bailed, who worked with the folks at Matrox on maintaining the drivers, also bailed, and they didn't find someone to replace him. This left Matrox in general, and 3D in Matrox in particular, without any support on either the corporate end or the volunteer end. But people still go around saying that Matrox cards have well-supported free drivers, including effective 3D drivers. It isn't so. -c
