Future features:
1. Object-oriented C++ interface + Python support.
2. Asynchronous memory access.
3. Eliminate if-statement in ft_sph
4. Culling of partially or fully aliased data. (already implemented using
CUDPP for compaction, but not tested).
5. Splatting - using CUDA if NVIDIA enables floating-point atomic
operations, using OpenGL with additive blending otherwise.
6. Hybrid FVR-Splatting. Use CUDPP for sorting.
7. Adaptive FVR: Sort terms based on frequency content and lowpass subsets
of terms as much as possible.
Original issue reported on code.google.com by
andrew.c...@gmail.comon 9 Apr 2008 at 2:58