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
@AndyGrimm
Here is a question I have. In the image you just posted, it still looks to me like there is quite a bit of translucency baked in on the nose and cheeks. If that image was used as an albedo map, it seems like the translucency would be too much in those areas if it was set correctly for the rest of the skin. This is why I have resorted to using the diffuse map in the translucency color channel for some texture sets. So I guess my question is whether it makes more sense to make some localized color adjustments to the diffuse texture to reduce the redness, and use a control map for translucency weight. Or... make a modified transluceny color map that takes the baked in red areas in to account?
Or am I making this way more complicated than is necessary?
In another thread somewhere, I recall @MEC4D saying that the diffuse texture should be almost a yellowish-grey color, and to let the translucency, SSS, and IOR define the actual color of the skin (for cuacasion skins). I have yet to get a successful result this way though, but that was before I started using the Spec/Gloss mixing.
Alright, so I guess I thought I had an idea as to how one might manage these colors, but I guess I was wrong.
I'm looking at a histogram right now and I see that the ranges are quite different.
How would I go about balancing this image to those ranges?
Phillip and Bree have Translucency weight textures in both M/V5 and BM/BF UV maps. I've not found anything similar anywhere else... which is annoying, because I'd prefer to have eyebrows managed separately, buuut...
They are pretty much my go-to for translucency, highlighting ears, nose, and fingertips. That said, you could probably make your own without too much difficulty...
I really dislike using Diffuse map for translucency color.
Yeah, Will... those maps were made for UberSurface. I know you swear by them, but I'm not so much of a fan as I once was.
But maybe I will try them again... the Interjection series are very similar.
Well, I'd happily use something BETTER if someone made them. Until then... ;)
Or you can just make them yourself -- translucency map isn't exactly a precise thing, so might be amenable to mitten-hands image editing. (I can't draw freehand to save my life)
Edit: If I were going to do my own maps, I'd probably just make them a dark gray, then put light blobs around the nostrils (the maps I mention are a little too light too high up the nose, IMO), ears, and fingertips.
Yeah, I can't say I'm a fan of using the diffuse map as the translucency color either. I've made custom weight maps as well, but sometimes the time it takes to make a good weight map just isn't worth it if it's not a skin I'm going to use regularly.
Unless you're a chronically anal retentive perfectionist such as myself.
And no, I can't draw freehand worth a flip either lol
I only use PBR SG base mixing and I use the same diffuse map for translucency. Diffuse maps are sRGB without speculars.
Should be...but are any but the newest ones (at least ones since Iray came out), for maps in store products?
Can't tell. But I find it easy to work with old (3DL) textures. The new ones are difficult to make them look good.
BTW. I'm 48 and a few years back found out I was capable of drawing without ever being in a arts school though I don't draw that often.
@KurzonDax
i prefer to study those projects which are available to the public which do research how to get the best possible results for human skin and expression...the example above is a diffuse/albedo real photo from the emily project at http://gl.ict.usc.edu/Research/DigitalEmily2/
Nothing against Mec4d's way - but the example texture which wikihuman made from the scans dosent look greyish to me. And their result is one of the best i ever saw.. Also - the redish problem is now solved from 4.8.. that means what before was grey can now be again more redish.
it does not mean that you must use exactly THIS range .... a tanned skin will be darker... important is to understand that all 3 colors are in a narrow range...
Thanks, Andy, makes sense. Yeah, the red issues in 4.8 just about drove me to drinking (at least more than I do already). I strongly suspect that issue in 4.8 is partly to blame for some of the less than great textures and shaders that have came out since the release of Iray.
I remember that, somewhere within the "Show Us Your Iray Renders. Part II" thread. The idea behind is to get a bloodless looking diffuse/albedo map, since you'll let let your Translucency Color (red) add the effect of the underlying "bloody" Dermis skin layer (while the Transmitted Color of the SSS parameter will add the melanin coloring effect of the Epidermis skin layer). The idea isn't that bad.
Unfortunately, someone seems to only have read the translucency part, put a red color into the Translucency Color slot with a fancy 4k translucency mask map, fiddeled in a ridiculous pink Transmitted Color, and while that turned out to add too much red, he/she/them using the SSS Reflectance Tint to draw out some of the too much red he/she/they added just a blink ago and call it a day. On texture maps which already have the translucency "built in".
Using the same common diffuse map for caucasian skin tones makes much more sense than anything else (except maybe using a melaninish color with a mask. Although you already have a certain level of that in your diffuse, too. But a bit more melanin effect doesn't hurt that much than too much haemoglobin effect). You need to have something there, but you can't use a red color since your diffuse map already contains it to a final amount. You already have a melaninish color at your common diffuse, and you need a mask to avoid your eyebrows/hairy texture part to get tinted by the translucency color... and to avoid your lips receiving a wrong melaninish color tone. They contain a lot more blood, and a "red" translucency would be more accurate there. In that cases, choosing a common diffuse for the translucence would be an acceptable choice.
btw i just test a texture which i matched to the values above.... i have clearly lesser problems with to hard contoures now.
my idea of a translucency map is based on obsevation .. thinner parts are more yellowish with backlight..(Ears)... thicker parts (nose from the side more redish, nostrils yellowish.).... lips red from blood...on the skull is not much to see...
Actually not so diffcult to paint one.
That pretty much sums up my early attempts at getting a quality skin. The whole time I was thinking this makes absolutely no bloody sense that I need to use the translucency tint to pull down the red hue. And with the red issue in v4.8, it was a spectacular excercise in futility.
@AndyGrimm the reddish in your skin textures are already backed in SSS information from the photo . Other programs use more channels for SSS than Iray so you can do better and skip some things.
Iray does not have all skin channels and you have do what the program offer you and researching Internet for clues will help you not much other than understanding better the process that you can't adopt to Iray in 100% in this case, as everything need to be controlled manually by the maps.
I still think that using diffuse maps for control translucency color is crazy not backed up idea , just to fake it until someone else make it.
Regarding tanned skin, no it will be not darker just have more yellow tone in it and the translucence level will dark it to the right value, only typical black skin will looks different more bluish charcoal and not brown, as brown is the mix of the SSS and skin , orange and bluish tones will give the brown with a drop of reddish in SSS
The SSS transiency undertone of the skin is always 255 red .. controlled by weight maps the desired ethnicity
and you can;t be against my presentation as it is only the way to find the balance , human skin is not red it is yellowish gray , it can have some purple-red discoloration as well but not red channel as that is the job of SSS, we can talk about until next decade, as I almost talk about for 2 decades , there is no magic or perfect settings for all ..as each skin textures need own unique weight maps and settings ..
so don't waste time for research as you will never adopt it to Iray the same way, you will miss too much crucial parameters
--Drop The Mic-
AFAIK, on DE 2 they don't use any translucency method, just Albedo, Specular and Subsurface Scattering. At least that's what I can see looking at the "emily_material.osl" file. I don't have Maya though, so can't say if it is different there.
Hopefully they just didn't tweak the renderering engine to fix the bonkers in their shader setup.
Mec4d, have you updated your iray skin shaders? Will you be releasing a new character any time soon?
@MEC4D welcome - i wondered myself why you never chime in this thread
...
Yes - Iray is a limitation...
I also dont think that a diffuse texturfor translucency e is the right thing - but i also dont think that 100% red is right....
We dont have a distance for translucency.. and lightwaves are filtered... a ear against a strong light well it is yellow! because more green and even blue wavelenght travel through... So.. the right way would be to paint a translucency map with simple.. some yellowish for thinner parts - reddish for thicker...
I think I'm going to experiment with no translucency and try to do it with refraction and SSS...
Not more or less crazy than to add reddish translucency to a texture map which already includes a final amount...
(Not aimed on your method though, just at the "standard" that most of your fellow PA's and DAZ itself follow these days).
Aaand yes, that's the idea behind it.
You need to have some translucency on translucent dielectics for the SSS to work properly. Else you won't have lightbeams shining through the "thin bits".
well all i can say - if in 4.8 the best way was to use a greyish texture... it is now in 4.9 a redish one...
It is true - i did tons of tests with greyish textures.. and my best renders had close to zero red and looked very pale in 4.8 - i finetuned the devibration way based on Mec4d.. now.. not anmore.. they look a lot like the example from wikihuman.
the difference is clear... when grey and 100% red did work in 4.8 - then i would need now 120% red in translucency or more red in the albedo
thinking about tanned or darker skin -> melanin stops light... there is lower translucency and sss.... = albedo is darker.!.. (and sss is more red - with a lower amount.
if somebody can correct me - please do so
Hmm.
Here's a test of IG Skin Essentials vs. refraction. I set refraction weight to .6 and used those translucence weight maps for refraction, also set transmitted color to a pale red and shortened range.
I find it interesting that the skin is actually 'drier' looking and more pale, except the more transparent bits. Also it ended up rendering in roughly the same amount of time.
I find this particular experiment kind of a wash -- both skins have something to say for them, in terms of realism.
(I realize that the skins are a little overtextured -- I'm using Macroskin which has a very aggressive Normal map that I usually lower to .3. Left it as is to see how the light would react)
(Although, reflecting on it, I should probably do a third image using the translucence weight maps for comparison)
@timmins.william...
i like the left one more... looks more alive