diff options
author | Sergey Sharybin <sergey.vfx@gmail.com> | 2016-02-17 17:10:02 +0300 |
---|---|---|
committer | Sergey Sharybin <sergey.vfx@gmail.com> | 2016-02-17 17:18:56 +0300 |
commit | 8cab3273163d49986943adba702a8ccc2094e230 (patch) | |
tree | 6be08e06df3bcc7438dc92782aea3e59a35d57f2 /intern/cycles/device/device_cuda.cpp | |
parent | d81dd2da0fc6c6a02c5484f800345b4af97dbb9f (diff) |
Cycles: Make CUDA 7.5 officially recommended
This was a hard decision, because going newer CUDA toolkit makes
rendering up to 5% slower. But on another hand, it solves major
speed regressions (up to 30%) with branched path tracing on a
top level cards.
Neither of those regressions have a meaningful and sane workaround
from the code itself.
Toolkit 6.5 could still be used, but it's no longer recommended one.
Diffstat (limited to 'intern/cycles/device/device_cuda.cpp')
-rw-r--r-- | intern/cycles/device/device_cuda.cpp | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/intern/cycles/device/device_cuda.cpp b/intern/cycles/device/device_cuda.cpp index 98997ae0968..80c8cb1e592 100644 --- a/intern/cycles/device/device_cuda.cpp +++ b/intern/cycles/device/device_cuda.cpp @@ -310,11 +310,11 @@ public: return ""; } if(cuda_version < 60) { - printf("Unsupported CUDA version %d.%d detected, you need CUDA 6.5.\n", cuda_version/10, cuda_version%10); + printf("Unsupported CUDA version %d.%d detected, you need CUDA 7.5.\n", cuda_version/10, cuda_version%10); return ""; } - else if(cuda_version != 65) - printf("CUDA version %d.%d detected, build may succeed but only CUDA 6.5 is officially supported.\n", cuda_version/10, cuda_version%10); + else if(cuda_version != 75) + printf("CUDA version %d.%d detected, build may succeed but only CUDA 7.5 is officially supported.\n", cuda_version/10, cuda_version%10); /* compile */ string kernel = path_join(kernel_path, path_join("kernels", path_join("cuda", "kernel.cu"))); |