NVIDIA SRAA Whitepaper and CUDA Implementation

NVIDIA SRAA whitepaper



NVIDIA has published a whitepaper and presentations slides (I3D 2011) about SRAA (Subpixel Reconstruction Anti-Aliasing).

Subpixel Reconstruction Antialiasing (SRAA) combines singlepixel (1X) shading with subpixel visibility to create antialiased images without increasing the shading cost. SRAA targets deferredshading renderers, which cannot use multisample antialiasing. SRAA operates as a post-process on a rendered image with superresolution depth and normal buffers, so it can be incorporated into an existing renderer without modifying the shaders. In this way SRAA resembles Morphological Antialiasing (MLAA), but the new algorithm can better respect geometric boundaries and has fixed runtime independent of scene and image complexity. SRAA benefits shading-bound applications. For example, our implementation evaluates SRAA in 1.8 ms (1280×720) to yield antialiasing quality comparable to 4-16X shading. Thus SRAA would produce a net speedup over supersampling for applications that spend 1 ms or more on shading; for comparison, most modern games spend 5-10ms shading. We also describe simplifications that increase performance by reducing quality.

You can download the whitepaper here: Subpixel Reconstruction Antialiasing @ NVIDIA

Here is the CUDA implementation of SRAA available at the end of the paper:

NVIDIA SRAA CUDA implementation

4 thoughts on “NVIDIA SRAA Whitepaper and CUDA Implementation”

  1. Promilus

    Well, it surely blurs image too, maybe less than MLAA, but still. And fences looks horrible! MLAA looks faster (but with lower quality), SSAA looks much better (but is much slower). Is SRAA the best of both worlds? I don’t think so, but again there was no implementation in drivers so no games tested with SRAA.

Comments are closed.