griefith/test
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Fix #131240: Cycles: Negative integration range in Huang Hair

Browse Source 
The cause is numerical issues with `fast_sinf()`. While fixing
`fast_sinf()` would ultimately fix the problem, it involves more
complications in other code paths, and it is safer to clamp the
integration range anyway.

Pull Request: https://projects.blender.org/blender/blender/pulls/131689
This commit is contained in:
Weizhen Huang
2024-12-10 21:56:04 +01:00
committed by Weizhen Huang
parent 2fd3a633db
commit a16879a5f0
1 changed files with 8 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
intern/cycles/kernel/closure/bsdf_principled_hair_huang.h
View File

@@ -856,6 +856,14 @@ ccl_device Spectrum bsdf_hair_huang_eval(KernelGlobals kg,
bsdf->extra->gamma_m_min = gamma_m_min;
bsdf->extra->gamma_m_max = gamma_m_max;
/* FIXME(weizhen): There is a problem of `fast_sinf(-1.57085085f)` returning 0 instead of
* expected -1, causing the range to be incorrect. As that seems an uncommon case, it feels safer
* to cover this case first, and look into solving the problem in `fast_sinf()` later. */
kernel_assert(bsdf->extra->gamma_m_min < bsdf->extra->gamma_m_max);
if (!(bsdf->extra->gamma_m_min < bsdf->extra->gamma_m_max)) {
return zero_spectrum();
}
const float projected_area = cos_theta(local_I) * dh;
return (bsdf_hair_huang_eval_r(kg, sc, local_I, local_O) +