News Archives

Khronos member Cadence Design Systems, and ArcSoft, announced they have partnered to develop AI and vision applications for Cadence Tensilica Vision DSPs. The OpenVX conformant Vision P6 DSP supports AI applications developed in the Caffe, TensorFlow and TensorFlowLite frameworks through the Tensilica Xtensa Neural Network Compiler, Android Neural Network API for on-device AI acceleration in Android-powered devices and includes complete, optimized support for more than 1,500 OpenCV-based vision and OpenVX 1.1 library functions.

Magic Leap released a handful of tutorials and assets files that will help developers get a head-start in creating mixed reality content on Magic Leap One. Magic Leap said that Unity and Unreal already offer optimizations for Magic Leap hardware. The headset has full support for OpenGL 4.5 and OpenGL ES 3.1, but Magic Leap recommends building applications with the Vulkan API for the best performance.

The Khronos Group announces the availability of the SYCL Adopters Program for the C++-based programming framework for parallel programming. Under the Adopters Program, implementers of SYCL 1.2.1 can access an extensive conformance test suite, and then upload their test results to Khronos for review and the opportunity to become officially conformant. Together with the SYCL Adopters Program, Khronos also announces the release of a maintenance update for SYCL 1.2.1, delivering specification clarifications that enable enhanced run-time optimizations.

Adam Sawicki, a member of AMD RTG’s Game Engineering team, has spent the best part of a year assisting one of the world’s biggest game studios in porting one of their AAA games to the Khronos Vulkan API. That kind of experience — embedded with the game developer and working hands-on in their codebase alongside their own engineers — is always worth sharing whenever possible. Adam has turned what he learned into a general presentation aimed at those looking to port a game engine to either Vulkan or DirectX 12.

Webinar: glTF 2.0: Status and Outlook - July 31In this Khronos Group glTF webinar, we will talk about the current status of glTF and its ecosystem, and why it is the "JPEG of 3D." We'll go into some of the current hot topics for glTF, and talk about what may be in the future for glTF. This webinar is appropriate for existing users of glTF 2.0 and those considering it as their 3D asset format. This webinar will be presented by one of the creators of the glTF standard and a member of the Khronos glTF Working Group. Be sure to register online.

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.

Khronos Safety Critical Advisory Forum: Details and Why You Should JoinIn April, Khronos introduced the Safety Critical Advisory Forum in response to developers’ growing concerns and demands of functional safety standards on hardware and software. The advice and support that the forum provides to Khronos Working Groups directly contributes to the creation of SC APIs. Members and non-members can contribute in the forum, this blog outlines the benefits of participation.

LunarG has released new Vulkan SDKs for Windows, Linux, and macOS based on the 1.1.77 header. Changes and additions to Vulkan SDK 1.1.77 include: Linux SDK is now packaged as a tar.gz file instead of a .run file; Many bug fixes, increased validation coverage and accuracy improvements, and feature additions and new extensions for this SDK release: VK_KHR_get_display_properties2 and VK_KHR_draw_indirect_count.

The Khronos Group NNEF Working Group Chair Peter McGuinness discusses fragmentation in the Machine Learning field. Machine learning capabilities are being added to everything from social media platforms, IoT devices and cameras to robots and cars. The pace of innovation is leading to fragmentation, and one potential consequence of that fragmentation is a risk of stalling. A universal transfer standard for neural networks will cut down time wasted on transfer and translation and provide a comprehensive, extensible and well-supported solution that all parts of the ecosystem can depend on. The Neural Network Exchange Format is one of two standards currently being developed to satisfy this need. Learn more about NNEF and how it aims to solve this issue.