Age | Commit message (Collapse) | Author |
|
|
|
|
|
|
|
Gives 6.5% speedup for hair.blend from testsuite.
This commit was previously reverted, but should work ok now. Patch by Sv. Lockal.
|
|
Also avoid some conditionals.
Reviewed by: brecht
Differential Revision: https://developer.blender.org/D310
|
|
|
|
material in cycles
Buggy MSVC 2008 in 32-bit mode ignores stack align attribute for float3.
Now it uses reference to __m128, which is always aligned.
|
|
yesterday.
KernelIntegrator just doesn't have valid data at this point, so we need to go one level deeper.
|
|
Reviewed by: brecht
Differential Revision: https://developer.blender.org/D306
|
|
Also fixed the redrawing while entering and exiting fullscreen, it would show
a distracting white window contents during the animation.
|
|
Indirect and Direct samples can now be clamped individually. This way we can clamp the indirect samples (fireflies), while keeping the direct highlights.
Example render: http://www.pasteall.org/pic/show.php?id=66586
WARNING: This breaks backwards compatibility. If you had Clamping enabled in an old file, you must re-enable either Direct/Indirect clamping or both again.
Reviewed by: brecht
Differential Revision: https://developer.blender.org/D303
|
|
Now these are removed from the curve.
|
|
|
|
in cycles
Also fix very similar problem in half-float SSE conversion.
|
|
See: http://clang-omp.github.io
+ fix a longstanding bad include in darwin-config
|
|
|
|
This way addons like network renderers can more easily reuse them.
Reviewed By: brecht
|
|
|
|
|
|
sampling.
|
|
importance sampling.
|
|
vector is not normalized.
|
|
Z, Index, normal, UV and vector passes are only affected by surfaces with alpha
transparency equal to or higher than this threshold. With value 0.0 the first
surface hit will always write to these passes, regardless of transparency. With
higher values surfaces that are mostly transparent can be skipped until an opaque
surface is encountered.
|
|
|
|
error handling
This patch adds a network_error() function more alike how other devices handle error's
- it adds a check for errors on load_kernels to make sure we do not crash if rendering without a server.
- it uses the non throwing variation of boost::asio::read.
Reviewers: brecht
Reviewed By: brecht
CC: brecht
Differential Revision: https://developer.blender.org/D86
|
|
|
|
Crash only happened when the number of openmp threads were greater than the resolution.
|
|
|
|
Gives 5% speedup for koro_final.blend, 10-15% for hair.blend
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D225
|
|
The AVX kernel functions for reading image textures could be get used from non-AVX
kernels. These are C++ class methods and need to be marked for inlining, all other
functions are static so they don't leak into other kernels.
|
|
|
|
Gives 0.7% - 1.3% speedup for BMW1M-MikePan scene.
Reviewers: juicyfruit
Differential Revision: https://developer.blender.org/D280
|
|
The same can be achieved by flipping normals on the mesh, but it can be
convenient to do this in the shader.
|
|
On earlier versions there is no proper multi-monitor support, so we leave the
choice still. But for 10.9 this just interacts better with other elements like
spaces and the dock.
|
|
This enables VEX-encoding in AVX kernel for windows msvc builds and gives 5-10% speedup for different scenes.
Reviewers: juicyfruit, dingto, brecht
Reviewed By: brecht
CC: brecht
Differential Revision: https://developer.blender.org/D284
|
|
Was a regression since avg_frame_rate changes.
Didn't find reliable way to get stream duration which will
work with both FFmpeg and Libav so added some freaking black
magic to distinguish one from another.
|
|
|
|
|
|
class, was never used.
|
|
coded, so define it in the kernel only.
|
|
|
|
Also move shuffle() in bvh traversal back (was accidentally lost in SSE hair revert)
|
|
|
|
|
|
nor vc2008 x64.
|
|
compilers that don't support it.
CMake still needs to updated to work the same for consistency, but this should
fix the OS X buildbot at least.
|
|
This is a known bug in Windows, now work around it.
https://bugreports.qt-project.org/browse/QTBUG-36192
http://connect.microsoft.com/VisualStudio/feedback/details/753224/regression-getsystemmetrics-delivers-different-values
|
|
This is my first stab at this and is based on this IRC converstation:
<mib2berlin> brecht: this is meaning as reminder only, I know you have other things to do > http://openvidia.sourceforge.net/index.php/Optimization_Notes#avoiding_busy_waits
<brecht> mib2berlin: thanks, bookmarked
only tested on Ubuntu 14.04 / cuda 5.0 but ill do some more testing tomorrow.
Also unsure about the placement and the lifetime of the stream and the event. But creating / deleting these seems to incur a non trivial cost.
Reviewers: brecht
Reviewed By: brecht
CC: mib2berlin, dingto
Differential Revision: https://developer.blender.org/D262
|
|
|
|
|