AMDs Erwin Couman presentation on accelerating game physics heterogeneous systems using OpenCL are available on Github for download. Sample videos showing rigid body work, broadphase benchmarks, gDEBugger and Bullet Physics library are available on his YouTube channel.
Open GPU invites original and innovative performance-oriented contributions from all areas of High Performance Computing (HPC) targeting hybrid GPU / CPU architectures, including hardware architectures, software tools or data structures. Topics can include, Compilation tools for GPU, New GPU oriented algorithm approaches, Parallel and distributed architectures, Modeling, OpenCL and CUDA performance analysis experiences, Hybrid architecture benchmarks, HPC languages, and more. Details are in an online PDF.
ReconstructMe is an easy to use real-time 3D reconstruction system. With ReconstructMe you simply grab your camera and move around freely. While you move, the system continuously updates and refines your model. Based on the idea of Kinect Fusion, we have improved various aspects and implemented the system in OpenCL.
This tutorial series from Rob Farber and The Code Project focuses on bringing knowledgeable C and C++ programmers quickly up to speed so they can work with OpenCL to write efficient portable parallel programs.
Post a comment describing your experience with reliability, stability and compatibility for any recent FirePro graphic cards in any OpenGL or OpenCL accelerated application. Posting enters you into a sweepstakes to win one of 3 AMD FirePro V5900 professional graphics cards. We are looking for feedback from professional users in the OpenGL and OpenCL community.
The demo version of a potential 3D Viewer for geo.admin.ch - the geo-portal of the Swiss Confederation - is online until July 31, 2012. The 3D Viewer is based on the open source globe "OpenWebGlobe" using WebGL. The scene covers entire Switzerland and is based on high resolution data from swisstopo: SWISSIMAGE, DHM25 and swissBOUNDARIES3D as well as extracts from swissBUILDINGS3D. In addition, textured 3D objects from FHNW Muttenz and EPFL Lausanne are available. Outside the area with high-resolution data, Blue Marble from NASA's Earth Observatory is used. The total amount of texture and geometry data exceeds 2 TB and is being handled using the cloud from Amazon Web Services. This demo has been developed by FHNW - University of Applied Sciences and Arts Northwestern Switzerland, camptocamp SA, the Federal Office of Topography, Switzerland, and geo.admin.ch
The Khronos Group is hosting a series of DevUs, educational sessions, Member F2F meetings and news conferences aimed at the educational community. The Khronos Pan Pacific Road Show 2012 is devoted to the Pan Pacific region with a focus on education and training, OpenGL and OpenGL ES, OpenCL, WebGL, WebCL OpenVG, OpenSL ES and OpenMAX. Complete details on all the upcoming events can be found on the Khronos Group event page. The next events will be held in Taipei on February 17 2012. We've just increased our number of seats, so registration is open again.
CAPS, the Many-core Programming Company, a leading provider of software and solutions for HPC Community will be offering two 2 day courses this March 13-14 and 15-16. The first course offers an introduction to parallelism with MPI and OpenMP. The second course will cover the basics of OpenCL. 50 computing hours on a hybrid cluster will be offered to each participant at no extra cost.
The newly announced AMD FirePro V3900 workstation graphics card is powered by the Turks GPU w/ 1GB of DDR3 memory, 480 stream processors and support for up to 5 display devices using AMD Eyefinity and DisplayPort 1.2. The GPU is designed to deliver OpenGL and OpenCL performance, reliability and compatibility of a high end graphics card at an entry level price point.
ArrayFire is a fast software library for GPU computing with an easy-to-use API for OpenCL and CUDA. Its array-based function set makes GPU programming simple. ArrayFire is available for C, C++, Fortran, and Python and integrates with AMD, Intel, and NVIDIA hardware. ArrayFire is currently available in beta form for AMD and Intel hardware using OpenCL.
ImageMagick now has certain algorithms which are OpenCL-enabled to take advantage of speed-ups offered by executing in concert across heterogeneous platforms consisting of CPUs, GPUs, and other processors.
The Khronos Group's growth continues to be significant and demonstrates the huge opportunity we share as members of this highly specialized industry. Throughout 2012, Khronos is hosting a series of media events, educational sessions and F2F meetings for consortium members and prospective clients. The Khronos Pan Pacific Road Show 2012 is devoted to our members and colleagues in the Pan Pacific region with a focus on education and training. Complete details on all the upcoming events can be found on the Khronos Group event page. The first events will be held in Taiwan starting February 13 2012.
CLOGS is a library for higher-level operations on top of the OpenCL C++ API. It is designed to integrate with other OpenCL code, including synchronization using OpenCL events.
Currently two operations are supported: radix sorting and exclusive scan. Radix sort supports all the unsigned integral types as keys, and all the built-in scalar and vector types suitable for storage in buffers as values. Scan supports all the integral types. It also supports vector types, which allows for limited multi-scan capabilities.