diff options
| author | Jeroen Bakker <jbakker> | 2019-11-13 10:03:44 +0300 |
|---|---|---|
| committer | Jeroen Bakker <j.bakker@atmind.nl> | 2019-11-13 14:55:44 +0300 |
| commit | e527544b766bb01fb95bff1d98fe5a246aaf7398 (patch) | |
| tree | 55389acb5f0bc87056dc9414c558231d870bf0ca /intern/cycles/kernel/svm/svm.h | |
| parent | ca1721270a6d3fa8d67b59a014ab94004e24026b (diff) | |
Cycles: OpenCL Performance
When using OpenCL with Cycles the rendering time increased substantial.
After doing some tests the bottleneck was found in 4d voronoi and 2d and
3d smooth voronoi.
This change will hide these behind a specific compile directive so the
speed will improve.
AMD RX480 + BMW scene
2.80 (3:10)
2.81 (5:48)
2.81 excluding 4d voronoi+2d/3d smooth (3:50)
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D6231
Diffstat (limited to 'intern/cycles/kernel/svm/svm.h')
| -rw-r--r-- | intern/cycles/kernel/svm/svm.h | 6 |
1 files changed, 0 insertions, 6 deletions
diff --git a/intern/cycles/kernel/svm/svm.h b/intern/cycles/kernel/svm/svm.h index 4fc5e633e3b..18f086d6726 100644 --- a/intern/cycles/kernel/svm/svm.h +++ b/intern/cycles/kernel/svm/svm.h @@ -214,9 +214,6 @@ CCL_NAMESPACE_END CCL_NAMESPACE_BEGIN -#define NODES_GROUP(group) ((group) <= __NODES_MAX_GROUP__) -#define NODES_FEATURE(feature) ((__NODES_FEATURES__ & (feature)) != 0) - /* Main Interpreter Loop */ ccl_device_noinline void svm_eval_nodes(KernelGlobals *kg, ShaderData *sd, @@ -545,9 +542,6 @@ ccl_device_noinline void svm_eval_nodes(KernelGlobals *kg, } } -#undef NODES_GROUP -#undef NODES_FEATURE - CCL_NAMESPACE_END #endif /* __SVM_H__ */ |