Khronos royalty-free open standards for 3D graphics, Virtual and Augmented Reality, Parallel Computing, Machine Learning, and Vision Processing

On the run up to next month's Shading Languages Symposium, keynote speaker Randi Rost has published this blog previewing his talk.

Lisuan Technology is a company dedicated to the research and development of graphics rendering GPUs. Founded by leading figures in the GPU industry, it possesses capabilities in large-scale chip R&D, top-level architecture design, software stack design, and mass production. They are interested in OpenCL, OpenGL, OpenGLES, Vulkan, WebGL, SPIR, and OpenXR.

LunarG announced the availability of Vulkan SDK 1.4.335. This release brings a long list of new extensions, improved validation tooling, clearer layer configuration, and – most notably – the first SDK-integrated version of KosmicKrisp, a Vulkan-over-Metal translation layer for Apple Silicon Macs.

Igalia's Ricardo Garcia pens this blog on the VK_EXT_cusom_resolve extension and it's role in the work Igalia has been doing to help Valve release the Steam Frame. 

VK_EXT_present_timing is conceptually simple yet powerful. The extension combines two fundamental features, which Vulkan devices can expose independently:

• The ability to receive timing feedback about previous presentation requests
• The ability to explicitly specify a target presentation time for each request

It is the combination of these features that enables applications to achieve smooth, consistent animation.

VIP9000NanoOi-FS provides hardware support for the OpenVX API and is compatible with a robust software stack and development tools, supporting OpenCL 3.0, OpenCL 1.2 Full Profile, OpenVX 1.3, OpenVX 1.2 Neural Network Extension, and extensions for OCL, OVX, and neural networks.

As part of VeriSilicon’s VIP9000 series, the VIP9000NanoOi-FS IP is designed for automotive and edge AI applications, providing high-performance neural network processing with a safety-compliant architecture. The IP supports a wide range of AI inferencing models including Large Language Models (LLMs) and Convolutional Neural Networks (CNNs), with hardware-accelerated AI inference, enabling real-time applications such as object and scene detection, augmented reality, robotics, face and gesture tracking, visual inspection systems, video surveillance, and advanced driver-assistance systems (ADAS). It natively supports multiple data types and features a programmable SIMD vision pipeline, enabling high-performance execution of neural network layers while maintaining the deterministic latency required by ASIL B functional safety standards.