Performance improvements: As with most Catalyst® releases, performance has increased in various situations. In this release of Catalyst® Intel 965 Crossfire performance increases due to improvements of the Crossfire D3D driver in handling large vertex meshes. Gains of 6.7-15.8% can be seen in 3DMark2006, 9.7-10.5% in 3DMark2005 and 3DMark2003 performance improves as much as 7.1%. . New Features HDR + Anti-Aliasing support for The Elder Scrolls Oblivion Users no longer need to download and install the Chuck patch – the Direct3D driver now includes native support for High Dynamic Range and Anti-Aliasing with The Elder Scrolls Oblivion. Official support for Stanford University’s Folding@Home GPU client As many of you know a few weeks ago ATI introduced stream computing to the world – i.e. using the power of the graphics processor for something other than rendering 3D graphics. The first major initiative of stream computing was a joint project with Stanford University called Folding@Home. Folding@Home uses the concept of distributed computing to use 100,000’s of PC systems to help simulate the process of folding proteins (many diseases/cancers are caused by proteins miss-folding). Stanford uses this data to try and better understand why proteins miss-fold and eventually work towards the goal of finding cures for diseases and cancers. On October 2nd Stanford released a GPU customized version of the Folding@Home client (we also released a beta of Catalyst 6.10, which is also required for the GPU Folding@Home client) that folds proteins 20-40 times faster than the CPU version of the client; please see our website for more details http://ati.amd.com/technology/streamcomputing/index.html Note that the GPU client requires a Radeon X1900/1950 series product in order to run. Like the CPU version of the Folding Client, the GPU client only runs when there are spare cycles available (i.e. when the CPU or GPU are sitting idle), so users don’t have to worry about the Folding client slowing down their systems when using it for work. We invite everyone to join the ATI folding team (team number 51394) regardless whether you’re using the CPU or GPU version of Stanford’s Folding@Home client. For those of you who already have a Folding team, you can ignore this point (unless you want to switch over to the ATI team