diff options
author | Brecht Van Lommel <brechtvanlommel@gmail.com> | 2019-06-18 19:30:48 +0300 |
---|---|---|
committer | Brecht Van Lommel <brechtvanlommel@gmail.com> | 2019-06-18 21:05:36 +0300 |
commit | b10921f0ccbd29f7f1fd19b9e9aa103b5a74fd9a (patch) | |
tree | 75445a729d00aa54156f024f857dc1ce3b11734a /intern/cycles/util/CMakeLists.txt | |
parent | 1dab26afb9c8f8e12c3e97a1503f09a4f437e77f (diff) |
Fix Cycles CUDA suboptimal performance on Windows 10 with recent graphics cards
When compute preemption is available we schedule more work which is more
efficient. However the CUDA driver appears to be incorrectly reporting this as
unavailable, even though it should be supported starting with Windows 10 1803
and Pascal and Turing (10x0 and 20x0) graphics cards.
This reduces render time by about a 25% difference on our benchmark scenes. On
Linux compute preemption appears to be reported correctly.
Diffstat (limited to 'intern/cycles/util/CMakeLists.txt')
-rw-r--r-- | intern/cycles/util/CMakeLists.txt | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/intern/cycles/util/CMakeLists.txt b/intern/cycles/util/CMakeLists.txt index 1c7a6549253..f0157b1be7b 100644 --- a/intern/cycles/util/CMakeLists.txt +++ b/intern/cycles/util/CMakeLists.txt @@ -25,6 +25,7 @@ set(SRC util_thread.cpp util_time.cpp util_transform.cpp + util_windows.cpp ) set(LIB |