archive/blender-archive
Archived
1
1
Fork 0
Code Pull Requests Packages Activity

Cycles: Add half precision float support for volumes with NanoVDB

Browse Source 
This patch makes it possible to change the precision with which to
store volume data in the NanoVDB data structure (as float, half, or
using variable bit quantization) via the previously unused precision
field in the volume data block.
It makes it possible to further reduce memory usage during
rendering, at a slight cost to the visual detail of a volume.

Differential Revision: https://developer.blender.org/D10023
This commit is contained in:
Patrick Mours
2022-05-20 18:01:26 +02:00
parent 14a5a91e0e
commit a8c81ffa83
16 changed files with 119 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
intern/cycles/device/cuda/device_impl.cpp
View File

@@ -1084,7 +1084,9 @@ void CUDADevice::tex_alloc(device_texture &mem)
need_texture_info = true;
if (mem.info.data_type != IMAGE_DATA_TYPE_NANOVDB_FLOAT &&
mem.info.data_type != IMAGE_DATA_TYPE_NANOVDB_FLOAT3) {
mem.info.data_type != IMAGE_DATA_TYPE_NANOVDB_FLOAT3 &&
mem.info.data_type != IMAGE_DATA_TYPE_NANOVDB_FPN &&
mem.info.data_type != IMAGE_DATA_TYPE_NANOVDB_FP16) {
CUDA_RESOURCE_DESC resDesc;
memset(&resDesc, 0, sizeof(resDesc));