Commit Graph

12 Commits

Author SHA1 Message Date
Brecht Van Lommel
e9d03296c7 Better fix for #36935 and 36316:
* 32 bit GCC builds now have the SSE BVH optimizations turned off, but still
  compile with SSE flags for better performance.

* White color when rendering on Windows seems to have been unrelated to SSE,
  rather it was a graphics driver not supporting half float textures, added a
  check for that now.
2013-10-05 19:56:34 +00:00
Brecht Van Lommel
29f6616d60 Cycles: viewport render now takes scene color management settings into account,
except for curves, that's still missing from the OpenColorIO GLSL shader.

The pixels are stored in a half float texture, converterd from full float with
native GPU instructions and SIMD on the CPU, so it should be pretty quick.
Using a GLSL shader is useful for GPU render because it avoids a copy through
CPU memory.
2013-08-30 23:49:38 +00:00
Brecht Van Lommel
01e22d1b9f Cycles: more code refactoring to rename things internally as well. Also change
property name back so we keep compatibility.
2013-08-23 14:34:34 +00:00
Brecht Van Lommel
b9ce231060 Cycles: relicense GNU GPL source code to Apache version 2.0.
More information in this post:
http://code.blender.org/

Thanks to all contributes for giving their permission!
2013-08-18 14:16:15 +00:00
Thomas Dinges
743a7a4a4b Cycles:
* GPU kernel can now be compiled without __NON_PROGRESSIVE__ again, was broken after my last commit. Also add a check for have_error(), in case the GPU kernel comes without Non-Progressive, to avoid a crash.

* Don't compile progressive kernel twice on CPU, if __NON_PROGRESSIVE__ would be disabled there.
2013-08-09 20:03:49 +00:00
Thomas Dinges
a18112249d Cycles / Non-Progressive integrator:
* Non-Progressive integrator is now available on the GPU (CUDA, sm_20 and above). 

Implementation details:
* kernel_path_trace() has been split up into two functions:
kernel_path_trace_non_progressive() and kernel_path_trace_progressive().

* We compile two CUDA kernel entry functions (in kernel.cu) for the two integrators, they are still inside one .cubin file but due to the kernel separation there should be no performance problem. I tested with the BMW file on my Geforce 540M and the render times were the same for 100 samples (1.57 min in my case).

This is part of my GSoC project, SVN merge of r59032 + manual merge of UI changes for this from my branch.
2013-08-09 18:47:25 +00:00
Brecht Van Lommel
7902fa57b6 Code cleanup: cycles
* Reshuffle SSE #ifdefs to try to avoid compilation errors enabling SSE on 32 bit.
* Remove CUDA kernel launch size exception on Mac, is not needed.
* Make OSL file compilation quiet like c/cpp files.
2013-06-26 23:29:33 +00:00
Thomas Dinges
372c4f80f4 * Comment out change from r57790 to fix compilation for now. 2013-06-26 23:15:52 +00:00
Thomas Dinges
15f5da4cd4 Cycles / SSE2:
* kernel_sse2 was built without actual SSE2 intrinsics on x86 systems.
2013-06-26 22:12:23 +00:00
Thomas Dinges
11707119de Cycles:
* Code cleanup, remove unused "resolution" variable from the DeviceTask class, was never used.
2013-05-14 21:18:20 +00:00
Thomas Dinges
13499c6240 Cycles UI Code:
* Some more variable cleanup for old texture output.
2013-02-04 18:50:09 +00:00
Brecht Van Lommel
7c9d993347 Fix cycles intersection issue with overlapping faces on windows 32 bit and CPU
without SSE3 support, due to 80 bit precision float register being used for one
bounding box but not the one next to it.
2013-02-04 16:12:37 +00:00