Artist friendly SSS

[Theverat]

See https://blenderartists.org/forum/showthread.php?448186-LuxCoreRender-2-0&p=3299997&viewfull=1#post3299997

The point I want to make with a good SSS implementation is that in Cycles, Arnold, Mental Ray, Renderman, and many other engines, you can control properties such as the density using a surface texture. Controlling the density in a true volume requires a volumetric texture, and there's no tool in Blender that makes making such a thing straightforward (outside of point density or smoke/flame effects). What would be nice is a volume that can be controlled using surface data (so you essentially get Random-Walk like scattering that would actually see light leave the mesh and illuminate nearby surfaces).

--- Want to back this issue? **[Post a bounty on it!](https://www.bountysource.com/issues/56760946-artist-friendly-sss?utm_campaign=plugin&utm_content=tracker%2F79505306&utm_medium=issues&utm_source=github)** We accept bounties via [Bountysource](https://www.bountysource.com/?utm_campaign=plugin&utm_content=tracker%2F79505306&utm_medium=issues&utm_source=github).

[Dade916]

I missing how this feature is supposed to be useful (and correct). It can be used to render what ?

And it is not correct from a physical point of view, a "color" of a scattering point is not be a property of the surface (?).

[Theverat]

My guess: It's useful if you can use surface texture like imagemaps to specify volume properties because it is very hard to edit volumes in most 3D applications.

Each point on the surface would probably define the volume properties in a certain radius with a falloff.

For example, you can specify the subsurface color with a surface texture in Cycles:

But I was just logging the recent feature requests when I made this issue. We can take our time looking into it, find out how other render engines handle SSS and what's the most user friendly way, and maybe collect some more opinions from other artists.

[Dade916]

This could be solved with some tool to automatically create heterogeneous volumes but the proposed solution is just plain wrong from physical point of view. I mean, it is not an approximation, it just produces a plain wrong result ^^