Page MenuHome

subsurface radius input bug
Open, Needs Triage by DeveloperPublic

Description

System Information
Operating system: Windows-10-10.0.17134 64 Bits
Graphics card: GeForce GTX 1070/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 431.70

Blender Version
Broken: version: 2.80 (sub 75), branch: master, commit date: 2019-07-29 14:47, hash: rBf6cb5f54494e
Worked: (optional)

Short description of error
the subsurface radius input is not overwritten by a vector input.

Exact steps for others to reproduce the error
[Please describe the exact steps needed to reproduce the issue]
video of the problem:

file here:

same problem happens with the sss shader's radius input also.

Details

Type
Bug

Event Timeline

this only happens with EEVEE by the way, in cycles the vector node works as expected.

This isn't a bug so much as it is a known limitation. From the manual Rendering » Eevee » Materials » Supported Nodes

Subsurface Scattering
Random Walk sampling is not supported. Per color channel Radius is specified by the default socket value. Any link plugged into this socket gets ignored. If Separate Albedo is off Texture Blur will be treated as always 1.0. Texture Blur is not accurate for any value other than 0.0 and 1.0.

I can confirm your results that a vector node plugged into the Subsurface Radius socket does not affect the amount of subsurface scattering.

Principled BSDF
Cumulate limitations from Diffuse BSDF, Glossy BSDF, Refraction BSDF and Subsurface Scattering. Anisotropy is not supported. Transmission Roughness is not supported. The Sheen layer is a crude approximation.

I think that the description here in the manual is somewhat incomplete. You have shown (and I can confirm) that a vector node plugged into the Subsurface Radius socket does affect the amount of subsurface scattering, just not in a predictable way. Perhaps this would be a better description for this report.

Here is more information on EEVEE's current limitations that you may wish to read.

Right, I believe this is indeed an Eevee limitation and not a bug per se.

@Clément Foucault (fclem) should be able to confirm.