AMD recently renamed their OpenCL Stream SDK to AMD Accelerated Parallel Processing (APP). APP is a set of advanced hardware and software technologies that enable AMD graphics processing cores (GPU), working in concert with the system’s x86 cores (CPU), to accelerate many applications beyond just graphics. This update also offers support for AMD E-Series and C-Series platforms (AMD Fusion APUs).
4pixels announces the first release of the CTK Workbench. The CTK Workbench is a powerful rapid prototyping environment. All kernels supported by the CTK can easily be connected to data sources and data viewers. New kernels can be created by entering the OpenCL source code and defining inputs and outputs. Images, video streams, binary data and audio streams are supported. All results are visible in real time.
AMD has released the FirePro v8.801 driver with full compatibility with OpenGL 4.1 on Windows and Linux. The unified driver is available for the entire current FirePro workstation graphics line (V3800, V4800, V5800, V7800, V8800 and V9800).
The Belgian scientist Johan Gielis discovered a mathematical formula to describe many shapes existing in nature. This formula is called the "Super Formula". To demonstrate GPGPU and to show that it can be used for many purposes including simply having some fun, we developed the "SuperShaper". A tool that uses OpenCL to create very high-detail 3D models using the Super Formula.
MATLAB announces the first release of OpenCL Toolbox. It is a simple interface that encapsulates clBuffer and clKernel as MATLAB class objects. It allows one to easily create device buffers, set device memory, compile and call kernel functions. MATLAB® is a high-level language and interactive environment that enables you to perform computationally intensive tasks faster than with traditional programming languages such as C, C++, and Fortran.
A focus on OpenCL API basics, covering the basic management using a small kernel as example. Lots of examples to help you get going quickly.
SiSoftware has posted two OpenCL benchmarks online. One addresses GPGPU OpenGL performance, and the second CPU OpenGL performance. The conclusion: There is no reason not to port CUDA code to OpenCL now!
NVIDIA announced a partnership with ARM -- code-named "Project Denver" -- to build ARM-based CPU cores that could power PCs, servers, and supercomputers.
The MOSIX group announces the release of the MOSIX Virtual OpenCL (VCL) cluster platform version 1.0, which allows OpenCL applications to transparently utilize many GPU devices in clusters.
Euro-Par is an annual series of international conferences dedicated to the promotion and advancement of all aspects of parallel and distributed computing. Euro-Par focuses on all aspects of hardware, software, algorithms and applications in this field. A call for papers has been put out with a deadline of January 24 2011 for abstracts and January 31 2011 for the full paper.