CuRast: Cuda-Based Software Rasterization for Billions of Triangles

TL;DR

CuRast introduces a CUDA-based software rasterizer capable of efficiently rendering billions of triangles without pre-constructed acceleration structures, outperforming Vulkan for dense, opaque meshes.

Contribution

It presents a novel 3-stage rasterization pipeline optimized for small triangles, enabling high-speed rendering of massive triangle datasets on GPUs.

Findings

CuRast achieves 2-5x faster rendering than Vulkan for large models.

It efficiently handles hundreds of millions of triangles in dense meshes.

Vulkan remains faster for low-poly meshes.

Abstract

Previous work shows that small triangles can be rasterized efficiently with compute shaders. Building on this insight, we explore how far this can be pushed for massive triangle datasets without the need to construct acceleration structures in advance. Method: A 3-stage rasterization pipeline first rasterizes small triangles directly in stage 1, using atomicMin to store the closest fragments. Larger triangles are forwarded to stages 2 and 3. Results: CuRast can render models with hundreds of millions of triangles up to 2-5x (unique) or up to 12x (instanced) faster than Vulkan. Vulkan remains an order of magnitude faster for low-poly meshes. Limitations: We currently focus on dense, opaque meshes that you would typically obtain from photogrammetry/3D reconstruction. Blending/Transparency is not yet supported, and scenes with thousands of low-poly meshes are not implemented efficiently. Future Work: To make it suitable for games and a wider range of use cases, future work will need to (1) optimize handling of scenes with tens of thousands of nodes/meshes, (2) add support for hierarchical clustered LODs such as those produced by Meshoptimizer, (3) add support for transparency, likely in its own stage so as to keep opaque rasterization untouched and fast. Source Code: https://github.com/m-schuetz/CuRast