[Bf-blender-cvs] [7765614eb3] id_override_static: Cycles: Speedup transparent shadows on CUDA

Sergey Sharybin noreply at git.blender.org
Thu Feb 9 14:52:31 CET 2017


Commit: 7765614eb39ffe7efdf42fc3e1b9b48b2b922e9f
Author: Sergey Sharybin
Date:   Wed Feb 8 13:05:05 2017 +0100
Branches: id_override_static
https://developer.blender.org/rB7765614eb39ffe7efdf42fc3e1b9b48b2b922e9f

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.

===================================================================

M	intern/cycles/kernel/kernel_types.h

===================================================================

diff --git a/intern/cycles/kernel/kernel_types.h b/intern/cycles/kernel/kernel_types.h
index 8c271c75e4..f518530106 100644
--- a/intern/cycles/kernel/kernel_types.h
+++ b/intern/cycles/kernel/kernel_types.h
@@ -84,6 +84,7 @@ CCL_NAMESPACE_BEGIN
 #  define __VOLUME_SCATTER__
 #  define __SUBSURFACE__
 #  define __CMJ__
+#  define __SHADOW_RECORD_ALL__
 #endif  /* __KERNEL_CUDA__ */
 
 #ifdef __KERNEL_OPENCL__




More information about the Bf-blender-cvs mailing list