I really wonder why GLSL shaders work faster than the fixed function ARB.. I'm really curious about this weird issue..
Because the current GLSL compiler in Mesa rocks and really produces optimized code. There's ongoing work to pass every shader (ARB assembly ones and fixed-function ones) through the GLSL compiler to optimize them a bit (and mainly to simplify things for hw drivers), but it's way harder to optimize low-level code than high-level one.
For me at least, on Natty Narwal, the r300g driver is preferable to the Catalyst driver. I get absolutely seamless multi-monitor support, including hotplugging ports on my laptop, on my docking bay, and the docking bay itself. I also get KMS, which is always awesome to have. I even get suspend and resume that always works.
I realize the open ATI/AMD drivers have lagged in the 3d speed realm, but they have also lacked in the multi-monitor/monitor hotplug/detection realm. I think in the latter they have come completely around to full support. For the former, it will be a slow process due to limited development dollars, but there will be progress. If it isn't fast enough progress, donate some cycles or dollars.
I'm not sure why people are complaining about video acceleration, I've never had a problem with it. My system spends most of its time at 800MHz while playing full screen video.
Well r300's are, what, 9 years old? The fact that they can provide a rich desktop experience is a great gift in itself. r600 and up should be concerned with high def video and full hd 3D.
I, for one, do not realy care that much about 3D games orr accel hd playback, as long as piwer saving, compositing 3D and stability rock. OpenCL would be a nice adterthought as in the next step. However it's floating point on the GPU and that is patented >