Friday, March 21st 2025
Microsoft DirectX Raytracing 1.2 and Neural Rendering Brings up to 10x Speedup for AMD, Intel, and NVIDIA GPUs
Microsoft's DirectX Raytracing (DXR) 1.2 announcement at GDC 2025 introduces two technical innovations that address fundamental ray tracing performance bottlenecks. Opacity micromaps (OMM) reduce the computational overhead in alpha-tested geometry by storing pre-computed opacity data, eliminating redundant ray-geometry intersection tests. Shader execution reordering (SER) tackles the inherent GPU inefficiency caused by incoherent ray behavior by dynamically grouping shader invocations with similar execution paths, minimizing thread divergence that has historically plagued ray tracing workloads. The real-world implications extend beyond Microsoft's claimed 2.3x OMM and 2x SER performance improvements. Both techniques are shifting development from brute-force computational approaches toward more intelligent resource management. Notably, both features require specific hardware support.
Hardware vendors' implementation timelines remain undefined despite NVIDIA's announced support across RTX GPUs, raising questions about broader ecosystem adoption rates. Microsoft's Shader Model 6.9 introduces cooperative vectors. This hardware acceleration architecture drastically improves matrix computation performance, enabling a 10x speedup in neural texture compression while reducing memory footprint by up to 75% compared to traditional methods. It bridges the gap between conventional rendering and neural rendering, with Intel, AMD, and NVIDIA already demonstrating implementations that combine path tracing with neural denoising algorithms, potentially making computationally intensive graphics accessible on mid-range consumer hardware by late 2025. While the technical merit of these advancements is clear, the April 2025 preview release timeline for the Agility SDK means developers face at least several months before these features can be meaningfully implemented in production environments.
Sources:
Microsoft, via Wccftech
Hardware vendors' implementation timelines remain undefined despite NVIDIA's announced support across RTX GPUs, raising questions about broader ecosystem adoption rates. Microsoft's Shader Model 6.9 introduces cooperative vectors. This hardware acceleration architecture drastically improves matrix computation performance, enabling a 10x speedup in neural texture compression while reducing memory footprint by up to 75% compared to traditional methods. It bridges the gap between conventional rendering and neural rendering, with Intel, AMD, and NVIDIA already demonstrating implementations that combine path tracing with neural denoising algorithms, potentially making computationally intensive graphics accessible on mid-range consumer hardware by late 2025. While the technical merit of these advancements is clear, the April 2025 preview release timeline for the Agility SDK means developers face at least several months before these features can be meaningfully implemented in production environments.
52 Comments on Microsoft DirectX Raytracing 1.2 and Neural Rendering Brings up to 10x Speedup for AMD, Intel, and NVIDIA GPUs
For Nvidia, Turing and up already support Opacity Micromaps (OMM), and Ada added support for Shader Execution Reordering (SER).
Intel says that Celestial will add support for OMM, and presumably it will support SER as well, but it wasn't clear on whether Alchemist and Battlemage will (or can) have SER.
AMD doesn't seem to support any of these on RDNA 2, 3, or 4, but may have their own scheduling tools that sort of mimic SER, but would require RDNA specific optimization (and why bother doing that when AMD is a tiny portion of the GPU market?)
Qualcomm will add OMM and SER support in their next generation.
For how much people LOVE to complain about "Ngreedia," it's once again obvious why they're essentially a gaming GPU monopoly.
DirectX do way more that graphics. It include DirectSound, DirectInput. It's a full blown gaming API where Vulkan is just graphics
In reality, the subcomponents of DirectX, Direct3D is the equivalent of Vulkan.
?!
OMM is also a proprietary feature of the NV driver.
Also, both of these features are used in about one single NV demonstrator - CP2077. No one else bothered to use them, and it's not because of their dx support...
SER and OMM were talked about when Ada launched, so it seems like nvidia beat AMD and Intel to the race and already have them, didn't hear anything about those from either AMD or Intel.So they will not optimize, maybe when AMD will claim more market share.
Do you mean developers won't optimize? With Nvidia having ~85-90% of the GPU market and already supporting these features on their most popular GPUs, there's a strong incentive to include these features for better performance, just like devs are finally getting comfortable requiring DX Ultimate features like ray tracing and mesh shaders now that most GPUs support it.
So we just got a new gen and it obviously hasn't got the hardware bits needed to do this with all the announced improvements.
Forever in beta. How are you guys enjoying the ride so far? I've got a few ICOs here and a bridge to sell, interested?
AMD has certain scheduling optimizations that may mimic how SER works, so if game developers take time to optimize for Radeon GPUs, the latter may get some speed improvements.
That's what I meant.
The Thing
I am just worried about the quality of Windows 11 and I am sure something is going to go wrong in some way.
That being said, can't we just transition away to Vulkan?
Microsoft DirectX Raytracing 1.2 and Neural Rendering Brings up to 10x Speedup for AMD, Intel, and NVIDIA GPUs"
Turn into thisAMD users need not apply. Maybe in 2027 AMD will release compatible hardware.
I think I may go back to reddit.