Sv. Lockal 20b046d763 Cycles: workaround for noise performance regression in CUDA 5.5 ... Use manual ternary operation widening in grad(). Without it nvcc 5.5 produces multiple branch splits with very big branches (because of inlining). This solves 19% performance regression for BMW1M-MikePan.blend. Also remove one redundant instruction in perlin SSE (when h == 12 or h == 14, then h is always >= 4). Reviewed By: brecht Differential Revision: https://developer.blender.org/D190		2014-01-08 22:25:55 +04:00
..
app	Cycles Volume Render: work on nodes and closures.	2013-12-28 16:57:02 +01:00
blender	Cycles: Fix World Homogeneous volume setting, had no influence.	2014-01-08 11:00:06 +01:00
bvh	spelling: use American spelling for canceled	2013-10-26 01:06:19 +00:00
cmake	Cycles / Standalone:	2013-08-30 17:34:27 +00:00
device	Code cleanup / Cycles: else/if for SSE41 kernel functions.	2014-01-06 03:22:14 +01:00
doc	Cycles: relicense GNU GPL source code to Apache version 2.0.	2013-08-18 14:16:15 +00:00
kernel	Cycles: workaround for noise performance regression in CUDA 5.5	2014-01-08 22:25:55 +04:00
render	Cycles Volume Render: scattering support.	2014-01-07 15:03:41 +01:00
subd	Cycles: experimental OpenSubdiv code.	2013-11-28 02:11:42 +01:00
util	Cycles: Don't include SIMD util file for OpenCL/CUDA, this fixes OpenCL compilation.	2014-01-06 21:28:18 +01:00
CMakeLists.txt	Cycles: experimental OpenSubdiv code.	2013-11-28 02:11:42 +01:00
SConscript	Added scons option to actually build the sse41 kernel. CMake already has it	2014-01-06 21:38:56 +01:00