Microsoft has released its latest Agility SDK, DirectX, which brings major ray tracing improvements with SER & OMM support.
Shader Execution Reordering & Opacity Micromaps Deliver Huge Improvements In Ray Tracing Performance: Now Added To Microsoft’s DirectX Agility SDK
Microsoft’s Agility SDK DirectX has received two new updates for the preview and retail releases, which support major features such as Shader Execution Reordering (SER) and Opacity Micromaps (OMM). The following are the major changes being introduced in each version of the Agility SDK:
Agility SDK 1.717-preview features
- Cooperative Vectors unlocks powerful new hardware acceleration for vector and matrix operations, making it easier for developers to integrate neural rendering techniques directly into real-time graphics pipelines with greater efficiency and performance. To learn more, check out our previous Cooperative Vectors blog post and GDC 2025 highlights, and stay tuned for a deep dive blog next week!
- Shader Execution Reordering (SER) is a new feature in DirectX Raytracing that allows applications to inform the driver how to reorder GPU threads for improved execution coherence and performance. By reducing divergence and improving parallelism, SER supported hardware can deliver a boost in performance up to 2x faster in path traced games. Learn more about SER in our full blog post or watch the GDC DirectX State of the Union Recording.
- Direct3D Video Encoding Updates include the following features:
- HEVC Reference List extension to support complex long-term reference (LTR) scenarios
- Two-pass encode with lower resolution 1st pass
- Adding Peak Signal to Noise Ratio (PSNR) to the encoded frame output stats
Agility SDK 1.616-retail features
- Opacity Micromaps (OMMs) is a new feature in DirectX Raytracing that takes advantage of hardware accelerated alpha testing during raytracing workloads. This reduced— and in some cases eliminates— AnyHit shader invocations, which boosts efficiency without sacrificing visual quality. In path traced games, OMMs can deliver performance improvements up to 2.3x faster by allowing supported hardware to more intelligently process complex transparency. Learn more about OMMs in our full blog post or watch the GDC DirectX State of the Union Recording.
- D3D12 Tiled Resource Tier 4 introduces support for creating tiled texture arrays with a full mip chain— removing a long-standing limitation that blocked packed mips in arrayed tiled resources. This unlocks more efficient texture streaming and greater layout flexibility for developers. See the full spec at Tiled Resource Tier 4. This release includes the following driver support:
- AMD: AMD driver support for Tiled Resource Tier 4 will be made available in early June 2025.
- Intel: Intel support for Tiled Resource Tier 4 is now available with the driver here.
- NVIDIA: NVIDIA will fully support this SDK release, please contact your developer relations representative for specifics.
For D3D12 Opacity Micromaps or OMM’s in short, Microsoft states that the new feature will enable the hardware to handle alpha-tested geometry much more efficiently than relying on costly AnyHit shared invocations.
In previous examples, Microsoft has touted up to 2.3x performance gains in path-traced titles for games that utilize OMM. One of the demos highlighted by NVIDIA shows over 60% gains. On the left, you can see the reference scene at 55 FPS, and the frame on the right has Opacity Micro-maps enabled, resulting in 90 FPS. Currently, only NVIDIA offers driver support for OMM for its RTX GPUs, while support from other vendors is coming in the future.
Alan Wake, from Remedy, is another game that relies heavily on alpha-tested geometry. A single scene can have anywhere from 4.5M + 9.3M triangles in total, with 5.2M skinned vertices and 2.2K skinned instances. The game is also one of the most intensive visual titles that utilizes path tracing, and with ray tracing set on high, the game can generate 10 rays per pixel or 36.9M per frame. An RTX 4090 without any optimizations takes 16.8ms on average to render a scene, but with SER and OMM, the frame can be rendered in just 10.2ms.
So, overall, some impressive gains with these next-gen technologies. Hopefully, we can expect to see the implementation of these features in newer titles so that developers can benefit from the latest hardware innovations embedded within modern GPUs.
