Page MenuHome

Principled Subsurface Scattering, wrong shadow color
Closed, ResolvedPublic


System Information
intel i5 6400, windows 10 64

Blender Version
blender 2.78.5 f3d9f0c

Short description of error

The color of the diffuse bounced light of Principled BSDF shaders with Subsurface Scattering is wrong. The color is always that of the Base color input, instead of the proper mixture of both Base color and Subsurface scattering color.

Both balls on left are materials using Principled BSDF, while the two on the right have better behaviour with a mixing of diffuse and subsurface scattering in the node editor. In all materials the diffuse is cyan (0,1,1) while the SSS is pure red (1,0,0)

In the bottom row diffuse and SSS are mixed 50%, which gives grey. The problem is that Principled BSDF gives shadows with only the color of the Base Color channel, in this case cyan, even though it should or grey

There's another problem that might be related. You can see that the bottom right grey ball, which uses a mix shader of diffuse and SSS at 50% has a red backscatter light, while the bottom left ball, the one with Principled BSDF and SSS also at 50%, has just a grey backscatter. It seems that, in Principled BSDF, the color of the SSS backscatter is the mixed color of both Base Color and SSS color inputs, instead of just the color of the SSS color input itself, which seems to defeat the purpose of SSS in the first place (although using SSS at 100% and controlling the radius of each color works fine, but it's just a single scatter) Tell me if I should report it in another bug report, or if it's not a bug, where can I discuss this behaviour with the apropiate developer.

Exact steps for others to reproduce the error

Event Timeline

I added a fix for this problem. It was just a small problem for diffuse ancestors. Here I was setting the subsurface value to 0.0 which led to a principled diffuse shader closure call. Here the base color was used to evaluate the closure. I just needed to apply the mixed color to base color in this case.

For your other issue, I added a new Diff (D2685) which corrects it.