We could design a very simple UI for each of the diffuse/transmittance/scattering components IF these wouldn't have to interact with each other to maintain realistic shading.
It might help if you think that these three light interactions always have to be in balance, they are not separate. It's not really a matter of thinSSS or SSS, even though they work very differently internally the concepts are the same.
Diffuse: Rough, smooth, dull, shiny. Easy.
Transmittance: Light that doesn't get reflected at the surface from the diffuse component, passes through the object and gets attenuated by the attenuation parameter, but also gets color and value (attenuation) from the transmittance color. A bit more difficult to understand but still pretty clear.
Scattering: Now there's a third interaction which is light that gets scattered inside the object, apart from light that is simply transmitted through the object without scattering. This scattered light may also take on a different color than the main Transmittance color you set. This happens in real life, for example in milk.
Each of these "parts" have one unit of light to work with and it has to be distributed based on your parameters for each of the parts - light reflected back from the surface, light transmitted, volumetric scattering of the transmitted light. Or in the case of thinSSS a faster, simpler non volumetric scattering model.
So with this in mind, going through your answers:
(I used the drapery material scene, I don't have the one you used, but tried the same material parameters. Also I don't say there aren't possible problems with thinSSS but the settings you used in your material would make it very difficult to check)
1. Why does lightening the Scattering color darken the material color?
Look at your asymmetry setting. It's set to a high negative value meaning almost all scattered light will be scattered in the opposite direction of the incoming light, so the "front" of the material in this case will have very little light reflected back towards you. If the sphere the cloth sits would be an emitter, you would instead see much brighter scattering (backlighting).
By brightening the scattering color you're also specifying more scattering, not just a color, so most of the light received by this material will be scattered, away from you, very little will be reflected back from the surface.
2. Why isn't this material white?
Again it's about the negative asymmetry setting, and the lighting/camera angle in this scene. You are specifying a material which sends most of the light through the back. Set asymmetry to 0 or even positive and it will be white. Such a high asymmetry setting may not be realistic and we have to do some testing probably.
But your current reflectance and transmittance values being so high, it's difficult to guess what happens when Maxwell tries to maintain the balance. You specify almost all light should be reflected from the surface, but also almost all light should be transmitted....
3. Why does Roughness have such a dramatic affect on how translucent the material is?
Because the light that gets diffusely reflected from the surface hides more and more the transparency/translucency.
This effect in your case is also more powerful because of the very low ND of 1.1 > ND really only matters when the roughness is low, as roughness increases the influence of the ND parameter decreases. An accurate ND can't be computed for rough surfaces. So when you increase roughness the refl0 color matters more and more, even with Force Fresnel on.
So in your case, with an ND of 1.1 and low roughness in the beginning, almost no light is reflected from the surface, instead it gets transmitted/scattered. The attenuation distance of 1mm is low, and also the asymmetry is negative so the light gets attenuated very quickly and also the light scatters away from the camera - a darker material.
4. Why does changing the Attenuation from 1mm to 1m have so little affect on translucency? Note: values under 1mm did make the object gradually more opaque as one would expect.
Asymmetry setting again. Even though you've specified now that the light should not be attenuated so quickly, most of it is scattered away from you.
5. What's the proper Nd for a leaf? Nd has a large impact on translucency but I'm not sure what an appropriate value is.
I don't know, but I'd keep it around 1.5, 1.4.
6. Why does Ref0 make such a large difference on translucency?
As mentioned above with roughness vs ND, since in that particular screenshot you have roughness set to 50, ND will not matter so much now and instead the amount of light reflected from the surface is determined more by the refl 0 color. The fact that light is diffusely scattered on the surface also hides the transparency/translucency of the material. The same thing would happen with a glass material.
To summarize - you've used the worse possible material settings for discovering how thinSSS works
A much better approach is to start as if you where creating a transparent material, no scattering. Set an appropriate ND, appropriate Transmittance (probably higher than 1mm even for a leaf) and use a transmittance texture if you want, but don't make the transmittance almost pure white. If you want a rougher surface you can also use this texture for the reflectance colors, but again don't just set them to white.
Then turn on scattering, it could be the same color as the transmittance for more consistent results and set a realistic thickness.
The asymmetry parameter is important and you can check its influence with the leaf preview scene to see how much backlighting you get by setting the asymmetry to a negative value.
The problem here might be that it looks ok now backlit, but when front lit, the surface is not solid looking enough. You can "cheat" here and create another layer with no transparency and apply your texture to the reflectance colors. Then play with the layer weight.
I type too much too......
- By Mark Bell
- By Mike Amos 20251018183343
- By Blanchett