Gpuopen News
The latest Gpuopen coverage curated for game developers.
AMD GPUOpen says ADLX 2.0 adds AI extensions and MCP servers so apps can monitor, manage, and optimize AMD graphics hardware. For game teams, that opens the door to agent-driven …
AMD GPUOpen details a one-step latent diffusion approach for real-time global illumination that aims to keep lighting stable without iterative denoising passes. For graphics programmers, the pitch is better GI …
AMD GPUOpen says its updated AMD FSR Unreal Engine plugin now supports Unreal Engine 5.8, bringing ML-powered upscaling and frame generation into the latest UE branch. The update also extends …
AMD GPUOpen details a lightweight attention-based neural GI model that reconstructs off-screen indirect illumination from reflective shadow maps. The pitch is practical for teams chasing better bounce lighting without a …
AMD GPUOpen details MiniDXNN v0.4.0, adding DirectX 12 SM 6.10 linear algebra support, input encodings, and neural texture compression. The update also ships a real-time GUI for training and visualizing …
AMD GPUOpen details a GPU work-graph approach for view-adaptive, crack-free subdivision of bicubic Bézier surfaces. The technique cuts triangle counts sharply while keeping output on par with hardware tessellation, which …
AMD has wired its GPU crash analysis pipeline into an LLM-friendly MCP server, letting agents turn a single prompt into post-mortem root-cause analysis and fix suggestions. For graphics teams, that …
RGP 2.7 is the headline here: AMD is finally exposing shader source alongside instruction-level divergence metrics, which should make wave-level debugging a lot less guessy. The update also adds Extended …
AMD’s RDNA 4 WMMA guide shows a practical way to do in-register matrix transpose without touching memory, using an identity-matrix trick. For graphics and compute programmers, that’s a nice latency …
AMD’s RDNA 4 WMMA guidance gets practical here: fuse the matrix ops so you double the K dimension and unlock 128-bit loads for FP8/INT8. The payoff is better memory bandwidth …