Which algorithms map best to GPUs and other vector-processors? What kind of algorithms are faster when using accelerators and OpenCL? StreamComputing investigates new problem spaces based on a starting point from “The 13 (computational) dwarfs of OpenCL” document by Professor Wu Feng and his group from VirginiaTech.
Altera announced that its SDK for OpenCL is conformant to the OpenCL 1.0 standard and is now included on the Khronos Group list of OpenCL conformant products. Altera is the only company to offer an FPGA-optimized OpenCL solution, allowing software developers to harness the massively parallel architecture of an FPGA for system acceleration. Altera will demonstrate its OpenCL solutions at the 2013 Linley Processor Conference, being held October 16-17 in Santa Clara, Calif.
Goo Technologies creates HTML 5 high-end graphics for games and interactive visualizations on the Web and is the company behind the Goo Engine. The findings of their 2013 State of Browser Gaming Index are out, and guess what, over half of americans play browser based games. This bodes well for WebGL. Goo Technologies is a web technology company whose aim is to make all digital experiences instantly available on all devices, everywhere using HTML5 and WebGL.
PCMark 8 now includes support for Windows 8.1 and some cleaned up OpenCL options.
AnandTech takes a look at Altera's OpenCL SDK for their FPGA. Altera introduced a private beta for OpenCL on FPGAs late last year and the SDK has now been made public. Altera's implementation is built on top of OpenCL 1.0 but offers custom extensions to tap into the unique features of FPGAs.
Vivante Corporation announced an area-optimized version of its GC series OpenGL 2.0 ES GPU cores for Wearable and other IoT devices. Featuring the industry's tiniest GPU core, less than one mm2, the GC400 gives MCU/MPU silicon vendors a simple plug-in solution and complete IoT software stack that easily adds full-featured smartphone graphics capabilities to any product line.
Join Intel for 3 live technical webinars with Q&A. Learn how to get the most performance out of visually intensive OpenCL applications using 4th Generation Intel Core processors with Intel Iris and Intel HD Graphics. Webinars in the series are: 1. Taking Advantage of Intel Graphics with OpenCL - November 6th - 11:00 AM PST; 2. Create, Build, and Debug OpenCL code with Intel SDK for OpenCL Applications - November 20th – 9:00 AM PST and 3. Analyzing OpenCL applications with Intel VTune Amplifier XE - December 4th – 9:00 AM PST. Register for the entire series of FREE webinars online.
Intel has announced that Release 2 (R2) of the Intel SDK for OpenCL Applications 2013 XE now adds Beta support for Microsoft Windows Operating Systems on Intel Xeon Phi coprocessors. The SDK continues to support Red Hat Linux and SUSE Linux. Available as a free download, the XE version provides a comprehensive development application environment that supports the full OpenCL 1.2 standard on platforms with Intel Xeon processors and Intel Xeon Phi coprocessors. The XE Version targets at developers of highly parallel applications including High Performance Compute (HPC), workstations, and data analytics. In addition, Intel provides an SDK designed for developers of visual computing applications. The Intel SDK for OpenCL Applications 2013 supports the Intel Iris Graphics and Intel HD Graphics family on 3rd and 4th Generation Intel Core Processors using Windows 7 and Windows 8 operating systems. Download the Intel SDK for OpenCL Applications 2013 XE Release 2.
Pico Pixel has added support for Khronos Texture Format (KTX). Also new is an SDK that makes it very easy to capture the color or depth buffer of any OpenGL program and send it to Pico Pixel for visualization.
The Khronos EGL Working Group is requesting feedback on a change to the definition of EGLint that would disallow using handles as attribute values. If you have feedback on this change, please post it in the discussion boards where the proposed change is described in more detail.