The landscape of APIs for accelerating vision and neural network software using specialized processors continues to rapidly evolve. Many industry-standard APIs, such as OpenCL and OpenVX, are being upgraded to increasingly focus on deep learning, and the industry is rapidly adopting the new generation of low-level, explicit GPU APIs, such as Vulkan, that tightly integrate graphics and compute. Neil Trevett presented the "Vision Acceleration API Landscape: Options and Trade-offs" tutorial at the May 2017 Embedded Vision Summit.
Amazon AppStream 2.0 is introducing Graphics Desktop and Graphics Pro instance families to deliver high performance graphics applications from AWS. The Graphics Desktop instance family offers a single instance type with an NVIDIA GPU based on K520 with 1,536 CUDA cores, 8 vCPUs, 15 GiB system memory, and 4 GiB graphics memory. This instance type is ideal for running desktop graphics applications such as Siemens NX, SolidWorks, ESRI ArcGIS, and other applications that use DirectX, OpenGL, OpenCL, and CUDA.
Intel has released their Graphics Driver 15.46 to provide launch support, as well as bugfixes, feature updates, Computer Vision and AI application development support, and support for Windows 10 Creator’s Update features. The driver is only for Windows 10 64-bit. Intel has enabled additional OpenCL media extensions, as well as a few preview extensions. In addition, 15.46 brings support for OpenGL v4.5, Vulkan v1.0.38, and programmable sample positions in Direct3D12.
ArrayFire announced the release of ArrayFire v3.5, an open source library of parallel computing functions supporting CUDA, OpenCL, and CPU devices. This new version of ArrayFire improves features and performance for applications in machine learning, computer vision, signal processing, statistics, finance, and more. Release notes are available and the source code can be found on Github.
In collaboration with Google, Codeplay is proud to announce the release of a new open-source tool allowing the compilation of OpenCL C language kernels to run on the Vulkan API. The tool, named 'clspv', allows a subset of the OpenCL C language to be targeted at the Vulkan API. This tool allows developers to port code containing more than a million lines of OpenCL C to run on the Vulkan API. The source is available on Github.
Synopsys announced that it has enhanced the convolutional neural network (CNN) engine in its DesignWare EV6x Vision Processors to address the increasing video resolution and frame rate requirements of high-performance embedded vision applications. To simplify software application development, the EV6x processors are supported by a comprehensive suite of tools and software. Combined with software development tools based on OpenVX, OpenCV and OpenCL C embedded vision standards, the MetaWare EV Development Toolkit offers a full suite of tools needed to accelerate embedded software development.
Apple announced several updates to the Mac lineup earlier this month at WWDC. Geekbench 4, which includes a new GPU Compute Benchmark that measures the performance of GPUs at performing compute tasks, shows that GPU performance with OpenCL has improved considerably with an increase of up to 80% when compared to the equivalent 2015 model. If you’re interested in how your computer compares you can download Geekbench 4. Find the complete benchmark results on the Geekbench website.
Duskborn Labs has just released part II of their OpenCL to Vulkan porting guide. Part I covers cl_platform_id -> VkInstance, cl_device_id -> VkPhysicalDevice and cl_context -> VkDevice. Part II covers porting from OpenCL’s cl_command_queue to Vulkan’s VkQueue.
Futuremark is launching PCMark 10, their seventh major update to the PCMark series of benchmarks first launched in 2002. PCMark 10 builds upon the PCMark 8 platform, adds a few workloads and streamlines the rest in order to present a vendor-neutral, complete, and easy-to-use benchmark for home and office environments. Anandtech has a nice review showing a little bit of OpenGL and OpenCL usage.
Codeplay has added OpenCL hardware support to Eigen, to offer a wider range of hardware to developers via the SYCL open standard. In this post, Codeplay talks about how they implemented the SYCL backend for Eigen to enable support for OpenCL hardware.