Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
OK. Finishing up a talk at Bill Moyer's web site about politics.... I'll get back to this in a few minutes. Thanks Andy!
OK. So far that's looking really good but.... I'm now seeing that unsightly glare back on the eyebrows. I have a mask for them but forgot which channel to plug that into along with the color if there is a color to set it too!
you set a eyebrow mask with BLACK eyebrows on WHITE background in:
translucency color
Glossiness color
Both channels, Translucency and Glossy color? Does look better. Funny, the pores are now looking less than before. So may have to go back in and turn those puppies back up to full strength in PS and resave out the bump map....
in my experience the bump values you had must be stronger for a realistic bump.. yes...
as said above.. you want to see them in specular highlights - this gives the " real" touch.. finaly.
Here is the finished render. This too almost 20 minutes. Not sure why..
nice :-)
now it goes tricky... because i dont know your diffuse map,,,
but try to make translucency color brighter.. let it like it is.. just brigther till the Red channel is about 220....Green and Blue you just overtake... you will get a brighther redish brown...
translucency weight about 0.45
then set transmission color to a grey -> 0.75, 0.75 0.75..
if she goes to pale for your taste make the translucency color darker...(only darker.)
Set SSS amount to about 0.5
And see if you like it...
A few months back I made a test with the Gemetry Shell (GS) approach, it is a diffusion test and compared my results against difussion profiles mentioned in the NVIDIA's GPU Gems 3 Book, "14.4.1 Diffusion Profiles", figure 14-9. Book mentions the use of laser beam but for the provided image I made a LED like object place on top of chin. They looked pretty similar.
Gotta run out for a while. You mean the Translucency color where I have the eyebrow mask plugged in?
yes your dark brown... just with the slider up till it is a brighter color .. i would test red 220 first....
@jag11
What exactly did you do?
the square is the light? a sphere with volume? .. i am not sure if i really understand your images.
Edit: Ahh.. on a chin? a very strong "led"... so we see the scattered light on a skin ?
Yep, put a little LED on top of of the chin of a female figure so you all see how light gets scattered with variations of using and not using Geometry Shell and also using different SSS direction values, positive and negative values.
Here they go, if you happened to have Scott 6 the better, note that glossy color map goes to glossy layered weight where it belongs.
Other than his eyes are looking a little flat (not reflection flat, but FLAT...as in not rounded), that is looking pretty darn good.
Interesting. Still, a transmission color of 1.0, 1.0, 1.0 is odd. The same amount of light reaching your GS reflektor will be reflected back to the surface. Actually, absorption and scattering go kinda hand-in-hand, and in tissues scattering is about 100-1000 times stronger than absorption.
Doing some tests with translucency and transmission/scattering in the DS 4.9 RC1. Found out so far that unless you set "Translucency Weight" to a full 1.0, the effects are very faint.
Unless you turn "Refraction Weight" to >0, "Refraction Index" does nothing. You could set it even to 5.0, nothing on your specular reflection will change. So you still have a "Glossy Specular" of the same as the default polyurethan plastic. Glossy Specular for a Refraction Index of 1.40 is a linear value of 0.028, 0.028, 0.028.
For "SSS Direction" I'd recommend a value between 0.81 and 0.9 (the latter is a "value typical for many tissues in the visible and NIR spectral ranges". Taken from the Bashkatov et al. research paper cited earlier in this thread).
OK. Back from dinner.
So then Translucency color is set to 0.75, 0.23, 0.23 and Transmitted Color is set to 0.75, 0.75, 0.75
Hope I understood that correctly. I couldn't find anything that's "transmission color" so I assumed it's Transmitted Color...
well.. it is acutally simply.. when Jaq sets 1.1.1. and uses a fully saturated albedomap including SSS allready backed in for translucency.. he gets a perfect surface which looks like scattered with the correct absorbation colors seetings...(just the opposite way - and very simple to set up).
i think the trick is cool - makes it easy for everybody to get good renders...
and whithout shell it looks good too... just set absorbation to a grey... 0.75, 0.75, 0.75... and you get lesser of the "effect"....
I always had a close to grey transmission color too.. just saying... i used a ratio from a wavelength penetration chart... and upped luminance...
yes - sorry.. transmitted color... and your bump map stronger :-)
So should Base Color Effect be set to "Scatter Only" or "Scatter and transmit"?
The Translucency color was way to high. Keep in mind she is a tanned Italian so these higher settings make her look very pale compared to how I envision her. Even with the Translucency set to 0.58, 0.13, 0.13 her skin looks not so realistic and a bit on the plastic side. Perhaps when I get the normals produced and all that it will look better. The quality looks OK to me but yea, there is something that's not quite realistic and toned to the proper shade in my eyes....
Thanks for the feedback, appreciated!
Rich
well she looks more and more like a italien to me
....
you mUST make the bump stronger.. the shine must get affected from the bump...tha'ts why she looks like smooth plastic..
and you can set translucency a little bit lower.... or go with the grey in transmitted color down.. 0.5 0.5 0.5
very close to a good one - just fine balancing now.. bump,glossiness (you can go little bit lower ).. and transmited color.
Some settings seem odd if we don't get the purpose, for example, a transmission color of 1.0 means no extra coloring, no extra tinting of light, inside the volume scattering means chaos, any ray of light that just enters the volume changes the color and direction before it bounces back to the surface. Most 1.0 values are used to indicate we want to use the texture as is...
scatter and transmit
From that point of view, for a "common" Diffuse Map use, that makes sense.
(I really didn't say that right now!...) 
But, you're aware of that your method sabotages the attempt to develop accurate physically based shaders and then force DAZ and it's PA's to create accurate textures following PBR standards; and now they've got a poor excuse not to, aren't you? *Traitor*
(Just joking.)
Huh? Show, or it doesn't exist and you're just saying things.
Even if you use a penetration chart to calculate an absorption coefficient for the different wavelength ranges to determine the corresponding transmission color, the outcome could never, ever, be a 'monochrome'. 
Jus' saying...
you set R G and Blue in transmitted color.... absorbation means FILTER... the SSS which is tranclucency color or map...
that's why mec4d picks a bright color from the face...(which is usally a yelowish beige.... very close to the wavelenght penetration ratio...... aka 3(4 red 1(4 g 1(/8 b.......
it is a single scatter because we have just ONE color.. but 3 channels...R G and B...
depending on the R G B ratio we filter the final SSS mix...
that's why 1 1 1 doess nothing iwith jacks map colors.. it leaves the colors as they are.. just BRIGHT..aka inifnit..1:1. maximum luminance for translucency
you can do this in 2 directions....
jacks eyeopner was that this works in two directions.. the maximum loss can be a little bit deepness.. because the 3 colors will not react with thickness (but we have just ONE thickness or his shell) - and because the LUMINANCE of the translucency colors is allready painted in his translucency map - you will not see a relevant difference...
Actually it should work even whithout a base color map... you can set the BUmp there as diffuse.
That's why Mec4d is right... with base texture is greyish..!
Just translucency can not be 100% red... because it is not possible to create a very pale skin color.which includes BLUE... once the luminance is 100% there is no more brightness.... no snow white skin with 100% red..
her best renders are all with dark and tanned skins because of that..
. otherwise .. lower translucency or pinkkish skin
The only trick is.. well render a slightly tanned skin in a very bright diffuse light and overexposure camera.. then yes
the Game PBR model shader - using 100% red .. drop some color information.. but look still good in 99%...as long as yellow and blue are in the base...
...thank you.
This is my first work with G3F. Josie7 makes for a real nice base for my several of my characters from my story, however, it will be some time before I transition over since none of the older generation skin maps are compatible and I cannot afford to purchase a bunch of G3 character morphs just to build a library of different skin maps.
Don't have PS and not up on creating my own skins. Really hoping that an updated version of Skin Builder Pro for both Iray and G3 is released soon.
you can only create blue and green.. from 100% red with reducing red..aka filter absorbation 0, 255, 255... that's why a snow white bright pale skin is not possible... because not enough luminance.
But Mec4d Method has the advantage that it is easy to dublicate and looks good for 99% of possible skin types.
... i hope it is clear that i just brainstormed with myself.. need to test it now 
My method setting a Brown - Beige .. works best.. because that's the real translucency color of humans.. BUT.. hard to do it right.. everytihng must be in the correct balance.. correct luminance and wavelength filter color....finnaly just a approximation too...
NOW jag's model.. well..Is actually a simplification of Emily...Humanwiki shader
there :
Greyish texture in single scatter (we have just base color but it does the trick...
Full Alebdo in Multiscatter - well we have a single.. we loose some scatter infos.. but most is still correct...
and that's it...
if i am right.. i dont need anymore a base texture.. just a greyish map there - a flattend bump for luminance in a shade of grey (simulating melanin absorbation)
Yes, I know that... ...absorption means actually something a bit more complex, but for easiness sakes and for Iray use, the Filter analogy is okay.
That penetration ratio is a bit crude, actually it's 750 µm (0.75 mm, at 700 nm), 230 µm (0.23 mm, at 500 nm), and 150 µm (0.15 mm, at 450 nm)... but anyways. 3/4 (0.75 red), 1/4 (0.25 green) and 1/8 (0.125 blue) isn't anything 'monochrome'.
Yep, we have 3 channels which make up the color. Ideally, to accurately determine the "Filter Value" for our transmission, we'd need the color equivalents for the wavelengths Violet (380 - 420 nm), Blue (420 - 490 nm), Green (490 - 575 nm), Yellow (575 - 585 nm), Orange (585 - 650) and Red (650 - 750 nm). And also ideally in about 10 nm increments. But as we already have been told and know: Iray is limited™. It offers only the Red, Green and Blue wavelenth ranges to determine our transmission color. 420 - 490 nm, 490 - 575 nm and 650 - 750 nm. That's the ranges we can pick a coefficent from to calculate our Transmitted Color value.
Ideally again, we'd need the same for our "SSS Amount", too, but for easiness sakes NVIDIA chose to use just a monochrome color there. (But I've been told they'll work on a change for that. No ETA mentioned, though).
[And that's why I said you still confuse absorption and scattering... "Transmitted Color" doesn't have anything to do with scattering. It`s absorption.]