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
@Rashad Carter
The settings are sort of my own aggregation of what I've learned here. For the diffuse base map, I did the modifications in Photoshop. For this particular texture set, I eneded up using a vibrance adjustment, then a saturation adjustment as a separate layer. Although the vibrance adjustment layer has a slider for saturation, I don't like the result when trying to do both in one.
The middle pic looks to me like it was rendered with an outdoor HDRI for the environmental lighting, not the slightly warmed spot lights I was using (5900K). And yes, the glowing nose is a bit of an issue. I actually had to paint in the shading inside the nostrils on the diffuse map to help tone it down to where it's at. It was full-on Rudolph with the original diffuse map.
Give me a few and I'll post screen shots of my settings.
Thanks indeed Kurzon!
Here are the settings I used. This one shows the negative SSSD. For the renders with the positive SSSD, I just changes it to 0.80.
And before I get grief, I know the glossy color isn't white. I actually forgot to set it back to white after an experiment, and didn't notice until I had done all of the renders for the images above.
Part of the problem is how Studio decides what to do with the map...the gamma correction feature in the Image Editor (not LIE, but the Image Editor you get when you click on the dropdown for each map...it's under Browse, etc). Basically, 0.00 means 'guess', by using the embedded information in the file itself. 1.00 means leave it as is and 2.20 (or any other numeric value) means apply the correction for THIS number. Iray is expecting one thing...but at 0.00 it may not be getting what it expects...especially with jpgs.
Just for an opinion of positive vs negative. In both cases, the angular repartition of energy of the SSS beam will cover the full solid angle (not only the half space), except if you opt for a pure +1 or -1. My feeling is that having a look at the way they separated/linked translucency and SSS in scattering, translucency has the main amount of energy in its "intensity diagram" in what would be the main propagation direction of light if it was simply refracted. Having positive SSS Direction will balance, IMHO, strongly scattering light repartition in the propagation direction, and reduce the amount of backscattered light (to the space light was coming from) via SSS. My preference after analysing front - back renders of simple figures with a backlight only or front and backlight would go to the negative directions. The equilibrium between transmitted light (scattered through) and back scattered light (SSS "reflected") seems more plausible for negative directions. This is just a preference regarding the way transmitted light due to scattering balanced with reflected light due to scattering.
The differences are not huge anyway. 0 SSSD will provide an isotropic repartition of SSS, positive ones will focuse more and more in the propagation direction (the transmitted direction), negative ones will focuse the energy lobe more and more in the "reflected" direction, but in all cases their will be a non negligeable part of SSS sent it the whole solid space. Just the balance front/back is different.
I forgot do you scatter transmit or scatter only?
@KurzonDax WOW - great improvments - your renders look now like mine - i cant make them a lot better from here...
the difference between plus and minus is in the shadow.... or that's where i try to improve since a while and that's why i use plus scatter....
the plus scatter -> where it must/should be imporved is in the ratio of transmitted distance and the ratio from red to green to blue....There is somehwere the SWEET spot where you get the more opaque look (nose) while still having the soft shadow with red fallow... THAT?S where i test right now....and very difficult to find the values.. see the test from Jcade and Jaq....
For real looking and "reacting" skin...... the base should not use full color range..... this migh be the difference which also your and other tests above show (the problems)...we want the DETAILS in a skin SHARP (diffuse)....while the scatter gives the deepnes and "Unsharpness" softness..... That's why it is basically wrong to use albedo AND translucency the same map...specially for a PALE skin... translucency will "oversample" the details....
the base map should have more GREY..... (i dont devibrate).. instead i SHIFT the dark red details (check.. moles, freckles skin lines...)... more in the black....and narrow the whole map in the used colorrange....
this is a OLD render serie.. but really ONLY g3f base.. showing theadvadvantage of this technique in Iray(shifting Black luminance)...split luminance from the colors in translucency..
NOTE that i always use DIFFICULT light and the same skinsettings.. that's where a real reacting skin DIFFERES from a made for a single shot render.... (now this old serie did not use plus scatter.. that's why shadows and countoures are a litle bit TO hard..... but no redish problems - real reacting to light (also colored)....and luminance....
AND the DETAILS in the skin are There!
With Studio v4.9 beta, I use scatter and transmit for the base color.
I think the reason I preferred the negative scattering on the ears is that it seems to better simulate the effects of the cartlidge in the ear by reducing the amount of light that is passing from the rear surface to the front surface. As you say, there is still some amount of light that scatters in both directions, regardless of the polarity of the direction value.
Unfortunately, if you set the ears to a negative direction, but the torso is set to positive, you end up with a seam that the average Studio user isn't going to want to spend the time on making adjustmens to eliminate. I love doing this stuff, and I'm not sure I want to spend the effort in trying to make it seamless!
@AndyGrimm
Probably my one biggest complaint with it right now is that some of the skin detail was lost. I get what you're saying about the oversampling, and suspect that is a contributing factor. I'm going to rework it some in photoshop later today and test again.
the most easy to do so is... add a layer - convert to black and white.... shift way more red in black and white..then G and B... then just overlay the layer with 50% opacity over the original base....
Narrowing down the BASE colorrange- comes from photos of REAL albedo skin... whitout Specular.... they never have full color range.... while Photos INCLUDING specular have a closer to full color range!...
that means base should have: Just a example...
R only in 20% aka 200 - 230.... Blue only in 150 - 180 .... and so on... and GREY (HSB B!)... B should be minimum 40 % !)
And translucency and Specular can bring in now all the missing colors and contrasts"...
The good renders all come close(in luminance and Black) to a reference photo...
.. It is mostly in the Black distribution aka Luminance !
THAT means: use this reference photo AND Measure R G B and Luminance and B....
Then do the same on your Render! Note the differences.. and work on them
PLus scatter does Solve the TO HARD look i got with real colormanagment in IRAY... that was my frustration since weeks - and why i could not Improve from what i had allready since 3 months
Here is my attemp at realistic skin. What do you think about it? Personally I don't like that much red in the nose area, I'll have to work that out.
i am agree with top coat - I dont use it! i said that allready on page 17 in this thread.. it gives the look of car paint instead of skin...
JUST switching in weight -> changes the look.....the only reason to use top coat on skin is for effects of makeup or to cheat fresnel values of skin..
Jaq - No i DID all my test with original textures - the work i do now since weeks - is searching and defining a clear replicable method based on : Rendering A FACE and not a skin.... using the stuff i KNOWN (colormixing, luminance, reaction on light, details, contoures)....for future or "how to made" textures.
Based on photo and print colormanagment AND the sience projects such as wikihuman - which also do it the same way - they USE the albedo in SSS... same as you do...but it does NOT have full color spectrum!
because that is what the render engines does for you - sss applies the missing colors !
@Rafmer
I think it IS a beautyfull PAINTING ! like from a old master -> Because.. the light information in the face does NOT match the sky.... there is so much NOT real in this image ->
But it does look GREAT...
different approaches - i try to create a FACE/skin.. where i dont care if the nose goes red - when the light MAKES it red - i want it red in ugly flash bulb light
Is a 0 SSS direction mean 'in all directions' or 'mainly perpendicular'?
the black face contour - chin and cheeks as example - is COMPLETE unreal in this light
.. on the chin it would be exactly oposite - a contour which goes BRIGHTER then the face.
Well I would say there is not priviledged direction for the SSS. So pretty isotropic. In my opinion.
Have a little time before heading out to work. Last night before bed I loaded up a fresh copy of Dawn, SP2 using the iRay settings for Tosca's maps she made for Dawn but switched out all my maps in their place, they all matched up to each section and as usual the Displacement wasn't used so that's cool so I seemed to be on the right track but low and behold.. you can plainly see that my maps are OK but there are some settings I'm not understanding.
The PBR type for Dawn 2 uses Metallicity/Roughness while Dawn 1 uses PBR Specular/Glossiness
I need to look over the settings closely to figure out what's going on.
Oh and Andy, I kinda figured that my Glossy map was all wrong! lol I was studying from another link, non DS, and saw these dark grainy maps.
Again I'm still catching up so I'll probably be driving everyone crazy until my old brain starts to get it. I apologize in advance!
Rich
PS. I was beginning to think it was my lighting but there is just one Distance light in the scene and that's it. So this is a setting in the surfaces tab!
@Rafmer if you like to make this image realistic...
THEN...
the sky you show is ether close to the blue hour or short or after the golden hour. low luminance (exposure.).
...
you must use COLORED light to match the hour.... A photographer would use the flash bulp just sparly.. but maybe bring in the blueish light with a reflector (there is blue sky to see.... this should color her left side a little bit...
Such sensible to see knowledge about photograpy and colormanagment - is the key for great renders ... and that's why i try to create a skin model which also DOES what i KNOW it should DO
Or it is a normal daylight sky - but the exposure is wrong set in the render settings.
With the settings I use I managed to get consistent results.
This is another one using Cailin from Raiya.
So got it figured out.
It seems that the Base setting for Scatter is correct for my needs and then enabling the Scatter and Transmitt under the Glossy and Top Coat areas is where that particular setting goes.
Seems much better.
When I get some time tomorrow I'll change the Glossy noisy map by inverting them as Andy suggested I try.
I would believe that this was the only problem, but whatever internal correction they're doing is pushing pure white brighter than pure white. This should not be possible with gamma correction.
(Also, for Pete's sake, the "default" whatever on earth adjustment they're doing should be the same as the one they apply to solid numeric colors, not some random different thing. Daz, please. Please. You're killing me.)
There is no way that blowout from dropping in a white strength map is working correctly.
Raiya Maps are in general very good, and this is always a pleasure to work on skin shaders with her maps. Nice results by the way! How is the bump map on this one? I would have increased bump just a little bit if the maps are good!... And now I'm gonna buy Cailin!
Thanks. I left the original bump map settings that came with the figure, 3.0, but I agree a little bit higher would work best.
Well, yes. The sky behind is just a backdrop; lightning is from the default HDR with some modifications to the burn white, crush black parameters, also a little desaturated. I'll post a more natural lightning for the next one.
Nice work in your last renders, by the way. I was skeptical about those hand made microsurface maps but they turned out quite good in that images.
Those look really, really good under that soft lighting. Are the settings essentially the same as your Scott 6 settings from several pages ago? I've been playing with those Scott 6 settings you posted, and liking the results.
I'd be curious to see how your Cailin material above looks with harsh, hard shadows like you might get by lighting with just one small spotlight.
Also, how are those settings looking with rim/edge lighting? In my skin experiments, I've felt the need to use an adjusted topcoat for my specular settings to get rim lighting to behave naturally like skin. I haven't tested it rigorously, but it feels like the fresnel effect of topcoat set to "fresnel" looks more like skin than the base glossy settings do. I could be totally mistaken about that though - has anyone made comparisons or have knowledge of how base layer glossy works vs. top coat glossy in regards to fresnel and lighting angle?
When I'm checking skin settings for consistency, these are kind of my basic tests - Testing to see if it looks good in soft light, in hard light, and if it has a skin-like response to edge/rim light.
well - all my renders which i uploaded in this thread do just show the "thing" which i try to improve at that moment. in other words i did not aim yet to make a perfect one ... because i known i must work on improved maps myself first... and it did not make sense to start to work on 16k maps before i know what i have to create in the iray sense
. This puzzle comes slowly together now.
what makes me "angry" is that jag's wrong approach using a shell solves all the transmitted distance and TCM problems - good to see in the shadow test some pages back...
but still - there is a lot missing also in his renders... the skin above is now TO soft and translucent... look just on the nose and you think it is a 4 years old face - upping the bump and lowering sss amount will change that.
but what i think i know now for sure - the current PA trend to BLUR the albedo for Base.. is the wrong approach. If you blur then ONLY the colors but NOT the greys! One must change and work in another colormodel to understand! RGB is a limited way to handle color AND luminance!
it starts from using RAW phototextures to start with - and ends with 32bit .exr textures if you want the best possible result. I think Iray can do it - the only limitation is that we can not change transmitted distance for each color... using a ratio in RGB TMC is more guessing than having control.
But maybe V3Digitimes can sheen more light on this - when her product is finally finish.