Reproduce “Dynamic GPU Voxel Grids and Voxel-sequential Processing in NanoVDB” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Eftychios Sifakis, Francis Williams, Christopher Horvath, Rohan Sawhney, Mark Harris, Jonathan Swartz, Ken Museth

The authors extend NanoVDB with GPU-native, bulk-synchronous topology updates and voxel-sequential access to active voxels. They separate topology from voxel data and rebuild the tree in coordinated epochs. On a 567-million-active-voxel flame-front grid, 26-connected dilation takes 2.0 ms, 183× faster than rebuilding with voxelsToGrid. This operator result is not a whole-simulation speedup; updates are not arbitrary per-voxel edits.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.