I thought this might be better here in it's own thread.

I'm not a fan of the complicated skin shaders that get bundled with products so here is my Cycles skin shader. It uses the Principled BSDF node for most of the lifting. It is not compatible with Firefly renders.

Here is the main shader. Skin maps sometimes need adjustment in tone. HSV node controls skin tone and saturation, Roughness and Clearcoat Roughness control the glossiness  of the skin and bump is controlled by the distance on the Bump2Normal node.

for lips I sometimes like to use a bit of metallic to simulate lipstick. Metallic is set to .5 and the roughness and Clearcoat roughness are increased by 1 in this example.

For the Face/Lips I like to use a mask to soften the hard line between the lips and face. The face and lips shaders are collapsed into compound nodes and mixed by a lip mask I made in Photoshop. This one is for V4. The edge where the lips and face mats meet is blurred a little to soften that edge. This combined Lip/Face shader is used on both the lips and the face zones.

for lips I sometimes like to use a bit of metallic to simulate lipstick. Metallic is set to .5 and the roughness and Clearcoat roughness are increased by 1 in this example.Also the HSV node was used here to lighten the color of the lips a bit. You can change the shade of lipstick used here without ever needing multiple maps for the lips.

For darker skins you probably will need to use a  smaller radius for the SSS. try .026, .008, .004. These numbers refer to the RGB channels and how deep the scattering is going for each color. 

That looks good, I particularly like the idea of using a small amount of metallic on the lips. 

these are the render settings I use

Ok, Just to prove a point. Your skin shader should look good in any lighting. Here are a bunch of examples. Same render settings and nothing was changed about the skin shader. Only the lighting has changed. In some cases I'm using only an HDRI on the background object and others I'm using the default Construct plus my own lighting sets

What unit of measure are you using in Poser?

Units is set to Feet. That might affect the SSS radius I suppose

What is 'redfilter' as a node please?

Nah, my Units are set in feet also.  Technically, the radius is supposed to be a measure in millimeters, the amount of penetration and reflectance of your RGB.  This is a separate calculation that is based on its relationship to the Subsurface scale.  Entering a Subsurface scale value of 1 would require you to use lower radius values; decrease the scale, increase the radius.  I normally use a scale value of 0.01-0.02 with radius values set to 0.48, 0.16, 0.109, for example.  It becomes a delicate balance between those two channels. 

The Poser13 documentation says that the SubSurface Radius is in millimeters

@Rhia474 That is the Cycles Converter node called SeparateRGB

@Y-Phil you are welcome!

What is 'redfilter' as a node please?

Xpost.

jinx

The Poser13 documentation says that the SubSurface Radius is in millimeters

Y-Phil posted at 4:50 PM Tue, 4 April 2023 - #4460779

The Poser13 documentation says that the SubSurface Radius is in millimeters

I never noticed this before in the manual unless it was recently added.  Based on this information and doing some quick dirty math, the values that I've been using are off.  New calculated numbers indicate a darker skin tone in Poser.  

For us who desperately try to follow this and understand/ replicate, explanations like above are much appreciated . :)

For us who desperately try to follow this and understand/ replicate, explanations like above are much appreciated . :)

This chart has been floating around for several years and has been the basic values used in both Blender and Poser.  I wonder if these values need to be reevaluated for renders.

Aha... Poser's Doc:

I could also be wrong with my settings. I was using a ratio of 7 to 2 to 1. Also I'm using the more accurate Random Walk SSS.

inner mouth parts

@ Y-Phil: I disagree with the Radius values, I think they're too low.

@ hborre not sure what the scaling is on the PBSDF node since it's not exposed for changing.

@ Y-Phil: I disagree with the Radius values, I think they're too low.

@ghostship2: The scale is similar to the Subsurface Strength, or Subsurface on the PBSDF.  At least, that's my understanding.

@hborre yes, I should RTFM lol

After doing some more renders with various skin maps I found an issue and a fix: some maps let too much SSS around the eye socket making them glow a bit. Here is the solution marked in red. Plug the Red channel from the SEPARATE RGB node into the SSS amount port along with the Clearcoat port.

@ghostship2

Looking great. Thanks.

May I suggest using the CombineRGB node plugged into the SubSurface Radius to control the tinting on the skin shader.  It's easier to input your values rather than typing a string of numbers.  

I also use it independently to check the color that the node is imparting so I get an idea of how the SSS is responding.  Below is an image showing the actual color being given in the shader arrangement above.

It's convenient to actually see what you're inputting rather than doing a trial-and-error guess of your radius values.

@hborre that's a great idea!

@hborre

I'm testing it out and it's doing funny stuff to the skin

Would your shader work with Poser 12 or is it just for P13?

Great shaders, by the way.

@hborre

I'm testing it out and it's doing funny stuff to the skin

@hborre

@jartz It should except that P12 doesn't have Random Walk for a SSS method in the PBSDF node or any other node. It's a superior method for SSS.

I think I know where the problem is.  I'm running test renders now.

Okay, thanks.

The recommended values added to the manual are based on recommendations for "diffuse mean path" from pixar's renderman (renderman.pixar.com) documentation. I believe those values are also behind the various presets in the FireFly Scatter node (Lighting > Special > Scatter)

During P13 development a lot of work was put into making sure that the various SSS nodes achieved consistent results with similar settings. The settings were added to the manual during all that testing, so yes, it is new information that was added for the P13 docs..

As regards to the "Scaling" value for Subsurface Scattering ... 

For Physical Surface Node, it would be the "ScatterScale" setting (fourth from bottom)

For Principled BSDF node, it would be the "Subsurface" value (fourth from the top)

It is also important to note that the scatter values entered in the radius DO NOT represent an RGB color. They are depth values.  In the example, RED scatters more deeply than GREEN, and BLUE is scattered less deeply than Red or Green.. 

So it makes it sound that implementation of a separate CombineRGB node, where P13 is concerned, is incorrect and that the Subsurface Radius values need to be inputted directly.  Also, there is a random walk and random walk fixed radius features under the SSS Method contextual menu, which one should be used?  

In addition, do underlying values, like the Surface Radius, become part of the multiplier calculations or do connector nodes override those values?  

I'll see if I can get Nerd to jump in here. I'm sure he can answer your questions far better than I can. 8-)

RE: Random Walk Fixed. According to Cycles documentation on that :

Provides accurate results for thin and curved objects. Random Walk uses true volumetric scattering inside the mesh, which means that it works best for closed meshes. Overlapping faces and holes in the mesh can cause problems.

Thank you for your assistance.  It's threads like this that help to establish an understanding of the software with the input of other users.  A big thank you to Ghostship2 for sharing your time and talent.  I also have to include Y-Phil, a source of knowledgeable information.

I've been thinking about this some. Let's say you have a really accurate skin texture that is highly photorealistic. The areas that are closer to bone (like the forehead) are more "yellow."  (IOW yellow undertones). Then you have blue veins that pop out. 

So, the color yellow is a mixture of R and G in the RGB color spectrum. The red values in the yellow color would be multiplied by the R depth, and the green value would be multiplied by the G depth.

And the Blue values in the veins would not scatter as deeply as the flesh colors.

 At least that's my understanding.But I pinged Nerd to jump in on this

Thank you for your assistance.  It's threads like this that help to establish an understanding of the software with the input of other users.  A big thank you to Ghostship2 for sharing your time and talent.  I also have to include Y-Phil, a source of knowledgeable information.

Ur... who? me? wow... 
Thank you