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
add MEC4D.. I learned a lot from you....
BUT.. you can take every skin penetration chart you want.... if you filter the LIGHT using a skinned animal or human body part photo based on datas... you will NOT get 100% red.. you will find that it looks Dark brown...(now you can play with luminance here too.. aka...
20, 30, 100 ... higher luminance.... up to maximum red .. you will also have more blue and green...(pale skin HAS a higher translucency luminance because light penetrates deeper).....
But i am FULLY aggreed - with you that at the end.. the total luminance is what makes it look right...(i do this by using a very NARROW band of RGB in Base)
And that we can use transmission color to set the final skin tone....
Actually my way does NOT differ much from yours....
@AbdyGrimm
Ok, that makes sense. I will have to play with this some later today. It may be a lead on why I've hit a bit of a wall with the grapes. I have translucency and refraction weight set to fairly high values.
Our skin models look SO HARD (like plastic).. because tranclucency does NOT get correctly scattered... only ONE color (transmission color) does this... and that's not enough to give the skin the soft deepnes look which we aim for.
And here we must start to cheat a little bit...(i do this by letting more SSS in the base... and use some backscattering which softens details on the skin...= Silk material you create with backscattering And anisotropy.. the same we can do to a small amount on skin.
ok... the old example.. why do Veins look BLUE.... and not red.... There it starts why translucency color can not be 100% in a model which does not inlcude correct light wavelentgh penetration.
A very pale skin has also MORE blue in the final SSS... how do we simulate this with 100% red? and a greyish base texture ?
Ok, saying that about Veins is just wrong... Arteries carry the oygenated blood, so they appear red, veins appear blue because the de-oxygenated blood appears blue. Not arguing with your point, just the term you used.
NO evilded Blood is ALWAYS red.. or did you ever bleed Blue?
,,,
The only difference is that blood in arteries have a very bright red... and veins a darker red...
Now a very pale (and translucent skin because of low melanin) skin does not look like pure red... because there is also a lot of green and red in the translucency...
Lets take a ALbino... NO melanin.... why does he not look100% red?
oxygenated blood is red.
because all wavelentghs can penetrate deeper and so come also more blue and green back to the the surface... actually close to the point that he looks snow white...but because of a more translucent skin you can also see his VEINS are red... (his eyes are red)
while you and me.. have blueish veins.
Can you tell me if this is normal.... for a Normal map. On my face textures I have it set to a full 1.00% but I had to set the rest of the body down to 0.25%. Weird.
Strenth with normal maps depends on how they were created. I know that most programs will let you adjust the strength of the normal map as your creating it. Is it possible that they were created with different settings?
Probably. I'll have to redo them when I'm finalizing things. I created them using the free on line creator, which I really like all the map options it gives you....
http://cpetry.github.io/NormalMap-Online/
here is a very PALE kid... i see a lot of yellow trough his years ... if we raise luminance on Brown (the real translucency color) we get a yellowish/redish yellow up to yellow/white,,,,
his ears are only red with a LOW backlight.. (there where only red wavelentgh can travel trough his ears.
Well... i explained a lot now why translucency color is BROWN and not red....and why very pale people start to look blueish in winter (check the kids face!)
The opposition can now explain why IT IS 100% RED in their conviction.
Yeah, I love Macroskin, but I find I almost always have to reduce the Normals to .3.
There was also a bug in the version of Iray included in DS 4.8 that affected SSS, I think at least some of the odd colour choices were compensating for that. If so they materials should look off in 4.9, which has a version of Iray in which the bug is fixed. Theer shoudl be updates for any affected sets.
Yeah, that's definitely true as well, and part of the reason I nearly gave of hope of creating a physically accurate shader prior to 4.9. What is odd though is that using a value of 1 for red in the transmitted color essentially exacerbates that bug by not allowing any red to be absorbed.
RAMWolff
because your normal map is made from a Bump map.. and Bump is resolution dependent.
if we use 4k resolution for the face... then the torso to be in the same scale needs to be about a 16k-32k file (but is also just 4k....so the grey range (128 = zero bump) must be 4 times lower in the torso bump before converting to normal. And some Pa's seems not to known that too.
in most converter you can set the SCALE (height) using a parameter.
Andy's right, I was oversimplifying and extrapolating. Blood is always red. Cyanosis can lead to a bluish or purpleish tint, but it is always red.
If you really want to design a skin like material in a PBR engine you only have to provide what the definition of the material needs, simple as that. Iray materials need an IOR, a surface color and if this is a translucent object the absoprtion and scattering amounts. There's no need to go cell size scale. If it doesn't look convincing it must be 'cause we are using wrong values and/or maybe a wrong texture.
An predominant mistake I see is the use of negative values for the scattering direction, a negative value means light returns to the source, 0 means no scattering and positive values means light follows its original path.
Other mistake I see is the use of transclucency color and transmitted color, unless you know what you're doing use just one or the other.
The image below is from my gallery, rendered in 4.9 in Hi-Res and not using top coat (hate top coat).
@jag11
..
very good one.....
i also see in my tests that translucency AND scattering (transmission color) seems not to work well together..(i showed the problem with translucency and refraction in a render before). But saying so in this forum would be close to be a revoluzzer
So i let this role to you.... you did not use translucency for this render?
I only use translucency color but transmitted color remains white as we only need scattering of translucency colored light.
ok... but does white as transmitted color not scatter? that would mean that every color R,G, and b - scatters infinit (no absorbation) i tought.. but you set a positive value.. so the light does not bounce to the surface? i see a lot of scatter in your render.. the softest nose i see yet in a close up.
Black 0.0.0 means zero transmitted color.,,
Oh man that looks awesome and the best yet, Please got a link to your gallery and what settings did you use to get this stage.
white color doesn't tint the already colored light inside the volume. You're right positive value means don't bounce light to the surface. Now you wonder. But how do you bounce it back? Using an opaque material. Works best than letting Iray do it alone.
right - i see allready that my model now finally has a red shine inside of the nose...
what do you mean with opaque? cutout opacity? or refraction index and weight ? or nothing at all? :-)
heck... i see it allready ,, nothing else... great.. why did you wait so long to show this render here?
Thanks both, my gallery is here, no big deal.
By opaque I mean an object not translucent, you can use a geometry shell or a skull like volume inside (.25mm) that reflects back the light, can have the same IOR. Ears look more convincing as back light travels and get colored inside the thin tissue.
As a side effect, you get faster renders 'cause scattering gets limited to .25mm and not the whole figure. That render was 3 mins to complete.
well it does the trick in my settings even without a inner shell.. the light gets bounced back from the inner backside of the full head mesh, .... works also fine with my narrow colorrange albedo and tranclucency colors which are outside of the colorrange from the albedo...
,,,, clear to me was only that light which shines trough translucency does not getting colored like it should using the recommended settings ...
now this approach is a eyeopener to me... i think i tried everything except setting transmission color to pure white and using plus scattering
Stuff for tests... thx
evilded and Kurzon: You got that pretty right. Welcome... to the Dark Side!
You still confuse scatter with absorption, Andy. That are two different effects. Scattering is only the vector change of a lightbeam. There's a publication available in our native tongue, which explains it as easy as very well.
You're right, 0, 0. 0 means no light will be transmitted through the volume (complete absorption), But 1.0, 1.0, 1.0 on the other hand means ALL light will be transmitted through the volume at a 100%.
Nice looking render, jag. But did you test your theory with different lighting settings and also a mid to large distance shot? If you place a bright spotlight behind your figure it will shine all the way through as if it wouldn't be there.
That geometry shell idea of yours sounds very interesting. I didn't work that much with geometry shells by now (to be honest, not that much like in "not at all" ).
By (.25mm) I assume you mean setting the "Offset Distance (cm)" to -0.25? Not sure if it would be worth all the effort, but with geometry shells you should be able to simulate all the different skin layers with their different attibutes. A method that MDL alone's not able to deliver that easily, since there you can't define layer strenghts in cm or mm. You also seem to have control over all the different material zones of human a figure, setting things to on or off. Interesting...
Wow. Must do some test renders as soon as I finish cleaning my monitor off from where my head just exploded.
I mean what @jag11 suggests and showed just completely unhinged everything I thought I had learned about Iray in the last eight months.
Hello @Jag11. Great render! Please, my english is not great.
Are you saying, that you created a skeleton with skull and bones under the skin to prevent light from transmitting too far in to the skin?