This sample uses the Cluster API available in OptiX 9.0 along with OpenSubdiv Lite. It runs on both Linux and Windows on most GPUs from Ampere onward. On recent hardware, we maintain a high frame rate even while generating millions of new micro-triangles every frame, for scenes like this:
The character here is animated, and we are re-tessellating and rebuilding the entire scene every frame using a camera adaptive tessellation algorithm. In this screenshot we are hitting 70 fps with 32.5M unique micro-triangles on a 4080.
Feel free to post comments and questions here on the forum and please report bugs as GitHub issues. Thanks!
update Jan 2026:
We updated this sample to use DLSS Ray Reconstruction via the CUDA API added to the DLSS SDK in version 310.5.3. You will need driver 590+ and at least an Ampere GPU (Ada or Blackwell preferred).
After doing a git pull or fresh clone, reconfigure using CMake and the DLSS dependencies will be automatically downloaded (see the README for other ways). Build and run the app and the denoiser should automatically be running; you can toggle it off or change quality modes using the Settings window. If you hit problems please report them, and if you’re really stuck you can still set ENABLE_DLSS=OFF in CMake as a workaround.
We are also planning a separate post to cover DLSS-RR in more detail now that it’s usable from CUDA directly!
Great repo! The linked tech blog article mentions that one can replace the user-space accumulation loop with DLSS Ray Reconstruction (DLSS-RR). It seems that DLSS should be integrated using the Streamline SDK which only works with DirectX/Vulkan (and only on Windows). Is there an undocumented way to use it (or at least DLSS-RR) in an OptiX program on Linux?
Thanks for the link! Last time I checked the NGX repository, Ray Reconstruction wasn’t available yet. Will try to integrate it with a pure OptiX pipeline but it would be amazing if NVIDIA open-sourced the OptiX implementation shown in this SIGGRAPH presentation. The NVIDIA-RTX/RTXMG repository only showcases DirectX/Vulkan integration.
Hey, sorry for the confusion here. I should have said that even though yes we’re thinking about it, a direct CUDA api for DLSS-RR is NOT currently enabled in public drivers and NGX dlls. I also replied to your post on the other forum.
Thanks for the clarification. I played around with the NGX SDK and modified the NVIDIA-RTX/RTXDI sample to use DLSS RR (Vulkan). My 2c regarding the OptiX integration:
The CUDA API for the DLSS SDK was released this week, and we updated the OptiX subd sample to use it!
Pull or re-clone the main branch and check the README. It’s set up to automatically download the DLSS dependencies during CMake config - you might see a short delay depending on your connection.
You’ll need Ampere+ and a 590+ driver. Hoping it just works for you.
Thanks for confirming. Yeah the default shading mode on optixSubd doesn’t even show it off that well. If you have the extra assets repo, some of the animated scenes like Barbarian work well with shading mode set to Base Color.
You seem like you’re pretty familiar with DLSS-RR already, but for people that aren’t, we might do a separate post down the road. The usage is different from the OptiX denoiser.
