Cycles: Speedup transparent shadows on CUDA
authorSergey Sharybin <sergey.vfx@gmail.com>
Wed, 8 Feb 2017 12:05:05 +0000 (13:05 +0100)
committerSergey Sharybin <sergey.vfx@gmail.com>
Wed, 8 Feb 2017 13:00:48 +0000 (14:00 +0100)
commit7447950bc3a3b11f0f1e0fd55df2031dbd3c0be2
tree0343fb115b3c62d21c8e83a79ea74248b9502e3d
parent9830eeb44b84cdb12e477cbb067f3b7e56634024
Cycles: Speedup transparent shadows on CUDA

This commit enables record-all behavior of transparent shadows
rays.

Render times difference goes as following:

               GTX 1080 render time
BMW                  -0.5%
Fishy Cat            -0.0%
Pabellon Barcelona   -11.6%
Classroom            +1.2%
Koro                 -58.6%

Kernel will now use some extra VRAM memory to store the intersection
array (200MB on my configuration). This we can optimize out with some
further commits.
intern/cycles/kernel/kernel_types.h