The OpenCL support by NNVM & TVM session from Linaro Connect 2018 in Hong Kong is now online. Abstract: To use mobile GPU to accelerate deep learning inference on ARM platforms in device side, OpenCL support seems a proper and promising fit. NNVM is an open compiler for AI frameworks with graph IR implementation, and TVM is an open source end-to-end Tensor IR/DSL stack. NNVM together with TVM provides a flexible architecture to support different frameworks and backends. OpenCL is one of the supported backends by NNVM & TVM now, the latest status and some how-tos will be discussed in this session.
The SHIELD Upgrade Experience 7.0 was rolled out on June 27th for all NVIDIA SHIELD TV devices, bringing with it support for Android 8 "Oreo" and Vulkan 1.1. If you don’t receive a push notification, you can manually check for updates under Settings > About > System Updates.
The X-Plane cross-platform flight simulator has been depending upon OpenGL for nearly two decades since the program first came into existence, but a port of its rendering engine to use the Vulkan API has been a work-in-progress. It looks like their Vulkan support is getting squared away as the company has tweeted this weekend they will be talking about Vulkan integration this weekend at the Flight Sim Expo in Las Vegas.
Synopsys, Inc. announced a new release of its DesignWare ARC MetaWare EV Development Toolkit. The ARC MetaWare EV Development Toolkit offers a programming environment for the EV6x vision processor IP based on the OpenVX open standard API, with a C/C++ compiler and OpenCL C vectorizing compiler. The Toolkit includes OpenVX kernels optimized to run on EV6x Processors, the first hardware-software platform to successfully pass The Khronos Group’s OpenVX 1.2 conformance tests.
The Khronos Group is working with Au-Zone Technologies to enable NNEF (Neural Network Exchange Format) files to be easily used with leading machine learning training frameworks. NNEF enables the optimized ingestion of trained neural networks into hardware inference engines on a diverse range of devices and platforms. Au-Zone is working with the Khronos NNEF Working Group to implement two purpose-built bidirectional converters, between TensorFlow and NNEF and also Caffe2 and NNEF. Both converters are expected to be released as open source projects to the development community in Q3 2018 under the Apache 2.0 license. Additionally, the NNEF and OpenVX Working Groups are working closely within Khronos to develop open-source importers, using the OpenVX Kernel Import extension, to enable the ingestion and execution of NNEF files.
If you were unable to attend IWOC 2018, you may be interest to hear that some of the videos are starting to appear online. First up is the "State of the OpenCL Nation" by Neil Trevett, President of The Khronos Group and VP at NVIDIA.
Compressonator is a set of tools that allows artists and developers to work easily with compressed assets and easily visualize the quality impact of various compression technologies. Compressonator allows users to choose from all major compression formats for their texture files and supports DirectX® 12, Vulkan® and OpenGL®. Users can view OBJ files in OpenGL and GLTF with DirectX 12. Version 3.1 will support viewing GLTF 2.0 files using a choice of OpenGL, Vulkan or DirectX 12.
What Game Developers need to know about The Khronos Group and the latest in GPU and 3D rendering for games from Neil Trevett, President of the Khronos Group. More presentations, video and photos from GDC 2018 can be viewed in the Khronos' Developer library.
The 2018.04 release of the Magnum C++11/C++14 graphics engine has its OpenGL wrapping layer optional, making it possible to use Vulkan or other custom renderers. There's now a first-class glTF format import together with general asset management improvements. A new example shows how to integrate Leap Motion hand tracking with Magnum.
FlextGL is a function loader generator with support for OpenGL, OpenGL ES and now also Vulkan. By selecting only a subset of features, extensions and functions that's actually needed it can significantly shrink down size of API headers for faster compilation times and smaller executable size.