The Folding@Home distributed computing program has long been able to run on GPUs, but the latest optimizations are for GPUs that support CUDA. As CUDA is a relatively closed source architecture from NVIDIA, that leaves AMD GPU users in the cold; with long time users still using CAL/Direct3D to power their GPU Folding efforts. According to a news story Hardware Canucks spotted, this is about to change as Stanford is working on optimizing F@H for OpenCL and the tests show almost a doubling of performance. Do keep in mind that F@H results can change from day to day but this is definitely a good sign.
"A few weeks ago, Stanford introduced the new Core 16 Project 11293 work units which are specifically tailored towards OpenCL-supporting AMD graphics cards. But do they bring the hoped-for increase in Folding@Home performance or is this yet another step towards disappointment?"
Here is some more Tech News from around the web:
- Leaked Intel Roadmap Reveals Sandy Bridge-E & Ivy Bridge @ [H]ard|OCP
- A Closer Look At Immersion Cooling For the Data Center @ Slashdot
- Nvidia releases technology enabling websites to stream stereoscopic 3D video @ DigiTimes
- AMD backs USB 3.0 on desktop and laptop chipsets @ The Register
- ASUS Releases Honeycomb-Based Eee Pad Transformer Source Code @ Linux.com
- Internet Explorer 10 Platform Preview 1: Hands On @ TechReviewSource
- AdBlock Plus man disputes Mozilla add-on tests @ The Register
- Beijing IDF 2011 – Day 2: Showcase and Demos @ Funky Kit
- Panasonic Media Convention 2011 @ t-break
- 2 Weeks Left to Win a BitFenix Colossus Window White Chassis @ eTeknix