Derek Gerstmann, University of Western Australia and Justin Hensley, AMD presented the course "OpenCL by Example" at SIGGRAPH ASIA 2010. All course notes and supplementary materials are now available online for download. The course presenters are experts on general-purpose GPU computation and advanced rendering from academia and industry, and have presented papers and tutorials on the topic at SIGGRAPH, Graphics Hardware, Supercomputing, and elsewhere. This course demonstrates how to combine traditional rendering APIs with advanced parallel computation using OpenCL (Open Computing Language), a cross-platform API for programming parallel systems such as GPUs.
In June 2011, developers, academics, industry leaders, and emerging innovators will convene in Bellevue, Washington at the inaugural AMD Fusion Developer Summit to learn about the AMD Fusion APU and the potential market opportunities it presents. AMD invites pioneers in heterogeneous computing to share their work and research findings in the form of presentations, technical papers, or reports with this influential community. Deadline for submissions is February 4th 2011.
The Khronos Group is hosting a Khronos Developer University co-located with SIGGRAPH Asia. Numerous Khronos members have joined together with each member taking a demonstration suite to create a large number of exciting demos in the Khronos Pavilion on the SIGGRAPH trade show floor. On December 17th Rightware is sponsoring a Beer and Demo Social. Complete dates, and schedules are available on the Khronos website. The Khronos Group has just wrapped up very successful events in Beijing and Tokyo.
The Khronos Group just wrapped up a DevU and press luncheon in Beijing. Presentations slides from all the sessions are now online. Some of the Khronos APIs covered at this DevU were OpenGL ES, OpenCL, OpenGL, OpenVG and OpenMAX and OpenSL ES by Vivante, ARM, Rightware, HUONE and NVIDIA. A complete schedule of events is available in our event archives.
A few advantages to using OpenCL on the CPU are that results are on the CPU for further processing, CPU-code can be mixed with OpenCL and specific algorithms can be processed faster because there is no PCIe-overhead. Streamcomputing has a closer look at at why you may find OpenCL on the CPU useful for your own work.
AMD to hand out free copies of their newly-published book "OpenCL programming on AMD CPU/GPUs" at the Khronos DevU on December 8th, and the December 9th Conference and Lunch. Errata: We had previously stated the book would be handed out at SIGGRAPH ASIA. In fact the book will be handed out at the Khronos DevU at Tsinghau University On December 8-9 2010.
CMSoft presents an easy-to-use powerful tool to compute the Discrete Fourier Transform using the Fast Fourier Transform algorithm with OpenCL, both single-precision and double-precision. Source code containing examples is available.
Two years ago Zack Rusin branched Mesa and called it Clover. Intended to provide OpenCL over Mesa, the work has been stalled for many months. A renewed interest in progressing work on Clover is now underway.
AMD just released Radeon HD 6000-series mobile model GPUs: the Radeon HD 6500M and Radeon HD 6300M. OpenGL 4.1 and OpenCL 1.1 support is included as well as DirectX compatibility.
NVIDIA has released their CUDA Toolkit 3.2. Lots of new goodness in this version, with special note the new OpenCL support. This means you can now use one toolkit for both CUDA and OpenCL. Support is currently only for Linux and Windows.