reduce_add()
with luminance()
to reduce color noise
weight > 0.0f
check due to floating point precision issue
reduce_add()
with luminance()
to reduce color noise
weight > 0.0f
check due to floating point precision issue
why isn't scale tilt applied at this level ?
The residual term accumulates all the contribution after TRT, and is assumed to be diffuse, so the tilt doesn't make a difference.
The result is not wrong, it just needs a few more samples to converge. Previously we deliberately give lobes with low contribution a minimal probability to sample, in case the next bounce has…
This is caused by 5b6f2ef5a26023c7e869a19fb8568c8faf8f2c79, mesh.normals_domain()
changed after split_faces
is defined.
But I don't understand why normals domain would change when switching…
I think the kernel is all C++ now, but not feature complete on some GPUs, not sure what exactly is supported but if buildbot doesn't complain it should be fine.
This does not compile on Metal. Add ccl_private
.