December 2011 saw the kick-off of an ambitious research project called “CARP: Correct and Efficient Accelerator Programming”, which aims to boost the programmability of accelerator hardware, such as graphics processing units (GPUs), by innovating in programming language design and implementation, as well as formal verification techniques. Funded by the European Commission’s Seventh Framework Programme (FP7), the consortium, which consists of eight partners--including Khronos members ARM, Imperial College London and Rightware--seeks to provide a unified flow for developing correct and efficient accelerator software, thus increasing reliability and energy efficiency of computing systems.
The AMD OpenCL APP SDK v2.7 now supports OpenCL 1.2 and improved C++ support for both host side and kernel side coding.
There will be five courses offered at Siggraph 2012 that relate to Khronos Technologies. Starting on August 5th followed by one course each day, there will be "Introduction to Modern OpenGL", "Virtual Texturing in Software and Hardware", "Beyond Programmable Shading", "GPU Shaders for OpenGL 4.x" and finally on August 9th "Graphics Programming for the Web." See all Khronos related events at Siggraph 2012 on the Khronos Group Siggraph event page.
Vivante Corporation today announced Vivante GC Cores passed the Khronos Group OpenCL 1.1 Embedded Profile (EP) conformance test suite on Freescale's i.MX 6 platform. The GC Cores use the latest programmable ScalarMorphic architecture to accelerate parallel data workloads on thousands of concurrent threads to achieve Gigaflops (GFLOPS) of computational performance. Applications taking advantage of Vivante cores to significantly speed-up processing includes image processing, computer vision, analytics, augmented reality and gesture-motion tracking.
Introduction to Modern OpenGL is an introductory course which offers an overview of the complete OpenGL pipeline, introducing all the latest shader stages. Other topics introduced are the shader-based pipeline, and a summary of key graphics concepts: the synthetic-camera model, transformations, viewing, and lighting. The instructors are Edward Angel, University of New Mexico and Dave Shreiner from ARM. Dave Shriener is co-author of the OpenGL® Programming Guide
Kishonti released GLBenchmark 2.5 Beta 3, a cross-platform graphics benchmark for OpenGL ES compliant mobile devices and embedded environments. Kishonti also makes CLBenchmark, a tool for comparing the computational performance of different platforms offering an unbiased way of testing and comparing the performance of implementations of OpenCL 1.1.
HiSilicon and ARM announced that HiSilicon has licensed a range of ARM Mali Graphics Processing Units (GPUs) including the market leading Mali-400 MP GPU and the latest high-performance Mali-T658 GPU. The new licenses will increase the scalability of the GPU performance points that HiSilicon will be able to offer manufacturers of mobile, consumer and home devices. HiSilicon has also licensed the latest ARM Cortex processor technology for use in next generation devices.
Sketchfab went live recently with new technology making it easy to publish interactive 3D content online. The free service does not require any third-party application installed on either the client side or the server. All Sketchfab needs is for the end-user to be running a WebGL compatible browser.
Imagination was showing off GPU compute on a cell phone chip at GDC, physics in your pocket. That demo was pretty simple, take a Pandaboard with a TI OMAP 4430, a dual-core ARM A9 CPU and an Imagination SGX540 GPU, and run a cloth simulation on it. Not only could the OpenCL version exploit the GPU to do more balls and sheets than the CPU version, but it saved power while doing so. How much? On one CPU, the simulation took about .68A@5V to run at 14FPS with 100% CPU load. With two A9 cores loaded, the Pandaboard pulled .84A and ran at 24FPS. In OpenCL, CPU load dropped to less than 30%, FPS jumped to 42, and power draw went down to .60A. More than 10% less energy used, 3x the frame rate, and you could run more simulations on the same box if you wanted. Not bad at all.