At the International Broadcast Conference, Vanguard Video announced H.265/HEVC codecs for a broad range of hardware and software platforms, including x86, ARM, OpenCL GPU, and FPGA IP Cores. Vanguard Video is previewing an OpenCL GPU HEVC encoder accelerator that more than doubles the performance of the V.265 software encoder running on the main processor. This technology was demonstrated at the Intel booth.
Imagination Technologies has developed a set of extensions to allow image conversion on the fly, using the Khronos Groups OpenCL and EGL technology. ImgTec converts the camera YUV data to RGB using the OES_EGL_image_external extension, and then use PowerVR’s zero-copy functionality to directly access the camera texture data from OpenCL. Find the in-depth explanation on ImgTec's website.
AMD announced that its collaboration with Adobe has reached a new milestone for more sophisticated professional video workflows. Using OpenCL acceleration means that video and film professionals, as well as enthusiasts, can take full advantage of the massively parallel compute power of AMD APUs, Radeon graphics, and AMD FirePro professional graphics when working with Adobe Premiere Pro CC, Adobe SpeedGrade CC, and Adobe Media Encoder CC by exploiting AMD App Acceleration.
Squid Systems announced today that leading broadcast equipment maker Harmonic and mobile software solutions leader NXP Software are demonstrating HEVC decoders at IBC 2013 based on the highly efficient Squid HEVC software decoder. Squid HEVC decoders run on CPU and CPU+GPU using OpenCL for platforms based on ARM Cortex A-Series cores and the Intel Atom (mobile) and high quality encoders on multi-core X86 processors.
AMD announced the HP ZBook 14 Mobile Workstation will feature AMD FirePro M4100 professional graphics, offering outstanding OpenCL performance and powered by AMD Graphics Core Next (GCN) architecture.
Xilinx is working with MathWorks and National Instruments to offer IP-based design environments for its Zynq system-on-chip FPGA devices. The design tools offer graphical and text-based programming languages such as C, C++, SystemC, and will soon support OpenCL.
AMD today announced its collaboration with Mixamo on the launch of Face Plus, an advanced real-time motion capture and 3D facial animation technology for the Unity game engine. Face Plus was developed for devices that support OpenCL™ 1.1 or newer versions.
Imagination Technologies has implemented a set of image and video handling extensions to the EGL Native Platform Interface to enable developers to create high-performance GPU compute applications on the new low-cost Hardkernel ODROID-XU development system. The new OpenCL API extensions enable developers to implement Instagram-like functionality on real-time camera data, including computational photography and video processing, while offloading the main CPU by providing resource sharing and interoperability between the camera pipeline, OpenGL ES graphics and OpenCL compute APIs. Imagination will be releasing the PowerVR GPU compute SDK and programming guidelines for PowerVR Series5XT GPUs and there is an upcoming release of the popular PowerVR Graphics SDK v3.2 featuring OpenCL support in PVRTrace for PowerVR Series6 GPUs.
This Khronos Group Chapter meetup will be held on Thursday 31st October 2013 and is one of the free sessions offered by the Mosaic3DX conference and exhibition this year. The Khronos UK Chapter is a group of the UK's Khronos API users enabling them to get together and expand understanding of the APIs, discuss ideas to improve them, and share technical achievements in the graphics/multimedia/compute domains. At Mosaic3DX this year, we would like to hear about your experiences developing games using APIs such as OpenGL, OpenGL ES, and OpenCL and thoughts on what you'd want to see in future versions. Expect some interesting presentations and a heated discussion!
PoCL is aiming at becoming an efficient open source (MIT-licensed) implementation of the OpenCL 1.2 (and soon OpenCL 2.0) standard. This release adds support for LLVM/Clang 3.3, employs inner loop parallelization in the kernel compiler, uses Vecmathlib for inlineable efficient math library implementations, contains plenty of bug fixes, and provides several new OpenCL API implementations.