vethanis, 14 days ago

@lisyarus Are you offsetting your ray start position by the (flat, non shaded) geometry normal * epsilon?

That's the usual solution to self shadowing acne artifacts at least.

For translucent materials you have to swap the sign of this bias based on the sign of dot(geomNormal, bsdfRay).

lisyarus, 14 days ago

@vethanis Yep, I'm doing that!

vethanis, 14 days ago

@lisyarus
Does changing the roughness improve or worsen the effect? If so, it would suggest precision loss somewhere in the BSDF.
Dividing by a variable less than 2^-23 is the usual suspect.

An rng bug could maybe introduce patterns, too, but most likely the BSDF.

I tend to reject sampling pdfs below 1e-6 since it will likely introduce either a firefly or +inf

lisyarus, 13 days ago

@vethanis It was the roughness, yep! I increased the minimal roughness value from 0.02 to 0.05 and it did the trick.