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
Running into an issue with the UberSurface. In your first skin example, for David 3 I've created all the desaturated maps and all that. Following along with UberSurface loaded (red teapot icon) I'm not finding a drop down with "skin4" ..... I've tried to load up UberSurface2 but it actually is removing just about all my options, down to like a diffuse color and a few other surfaces, not adding to the UberSurface default. So not sure what's going on with that. Any hints about getting that straightened out. Not sure why when I'm following along with tutorials I run into weird stuff like this... a script not working or a surface shader not showing me what I need to continue on. Grrrrrrrrrrrrrrrrrr
Thank you.
Off to dinner...
Rich
So I tried something last night before bed. I tried loading up the UberSurface 2 in 4.9. So not sure what version of DAZ Studio US2 originally supported but I can't use your tutorials for 4.5 apparently!
Have you tried reinstalling US2? If i remember correctly, the files don't install themselves to the runtime, but the actual DS "Program Files" folder (assuming you're using Windows). This is generally the case when you have a stable/public build and a beta build.
Hi,
It's a moot point. I went ahead and set up the body skin back to default and then tried the posing tools. Well 4.9 is far and away better. While the symmetry issue will probably persist and I'll have to get out my trusty Edit Pad Pro to double check values (or just do my poses in Poser and load them up via the Poser area from within DAZ Studio and save them out as .duf files, 4.5 sucks. There are allot of options missing that they added later on so yea, 4.5 is back of the bus again. I will probably uninstall it. So UberSurface DOES load fine in 4.9 so I don't think it was written for 4.5.
The hair looks especially awesome. Is the hair model subdivided or as-is?
So first of all, if we're talking about the recipe on p.31, then it's for the UberSurface2. If you look closely at the heading, you'll see that the skin shader is said to be US2, and only sclera is the original UberSurface.
Then, "skin4" is not a dropdown but a preset from the US2 folder. It modifies the scatter and absorption colour swatches and depths. There's more information about these two on p.19, and of course more tips in the whole section 2 (pp. 27-30).
If you want to start with David, I suggest that as a practical exercise, you adjust that recipe to work with "gamma correction" on and set to the correct "2.2" curve. Linear workflow is the only correct workflow =) ...most of those recipes were rendered using "Fantasy Lights" with the "Portrait Add-On", BTW.
And finally, there is a script called "UberSurface2 Upgrade" in the US2 folder that should be used when switching from the original UberSurface to UberSurface2. It will retain slider values etc.
Yes, I understand all that but I guess US2 wasn't written for 4.5. Must have come around 4.7 or something because it doesn't function nor load or "reveal" any other surfaces and in fact removes most of them. So aside from being discouraged by the older posing symmetry and save dialogs (You really don't know how great improvements are until you miss settings your used to having and use all the time) so I've basically put DS 4.5 away and am back to using 4.9 and of course these shaders work great under that version and I've gotten my skin looking like I wanted for the most part. Now I'm working on the iRay skin for the character, after that then into Poser to make presets there. YIKES!
Thank you!
To be honest, US2 predates the whole DS4.x line; a lot of the images in the SSS handbook were done in DS3. What you describe sounds very much like an install problem. The shader support files install to both "public build" and the main one. DIM should figure it out on its own, but if it doesn't, it's a trivial fix, you just need an admin user (supposing you install in the Program Files protected folder). Those support files haven't changed much (if at all) between DS versions.
Hmm, didn't know that. But I am the Admin of my system and have tried hand installing them as well as DIM installing them. no go...
As is.
Portia Hair by AprilYSH comes with a base res prop and a SubD one. Technically, you'd want to use SubD to smoothen things out where the strand (or more aptly, strips) bends and curves. I didn't use SubD cause I want to make sure the hair shader presets looks good with and without SubD.
One good idea I've implemented in the preset is to do displacement on the 2nd US2 channel. By doing so, you can change tiling for displacement but keep the opacity since the opacity is in the first channel. Of course, you need to restart the render to see displacement changes, but that's out of my hands.
I'm pretty satisfied with the darker shades but I had to make a compromise for blondes. Each hair prop behaves differently with the same preset (due to number of layers used, probably) Hair with more layers gets darker shading (from the ambient occlusion), while those with very few layers don't.
Re-install using 'Customize Installation'. It should bring up an extra dialog where you want it to install too. My guess is that by default it install to the most recent build (DS4.9 in your case), but not to the older build. It is the logical behavior.
Hmmm, but as stated the US2 works fine under 4.9 just not 4.5 so not understanding if US2 works fine under one and not the other?
I generally see the same problem when using stable builds and beta builds. Since US2 doesn't come as default shaders, the 'other' build doesn't know it exist. Objects/surfaces with US2 applied in the 'other' build will either render white or black. The solutions is always to reinstall it. To be honest though, I've never tried running two stable builds although I have downgraded to 4.7 a couple of times.
Basically, reinstalling places the .sdl files in the appropriate folders ie. C:\Program Files\DAZ 3D\DAZStudio4\shaders\omnifreaker\surface
And for the beta, the .sdl should be in C:\Program Files\DAZ 3D\DAZStudio4 Public Build\shaders\omnifreaker\surface
And also the relevant subfolders in these folders:
C:\Program Files\DAZ 3D\DAZStudio4\scripts\support\omnifreaker
C:\Program Files\DAZ 3D\DAZStudio4 Public Build\scripts\support\omnifreaker
should both contain the *Surface2*.dsa scripts. These link the shader to the DS interface.
Great idea =)
True.
Back to some UE2 bounce testing. A portal light scenario, with all lights placed outside. Used my faux area lights, with about half the energy intensity of PhysLights. Render options remains my preferred settings (max ray depth 12, shadow samples 16, diffuse bounces 4). Haven't found a setting for DelightGIHDRI yet, so didn't test it.
AO. 10 min 28.58 sec
Bounce. 18 min 39.82 sec
Looks pretty nice, though I think I used too high the bounce light intensities. This is with specular bounces limited to 3, hence they don't show all the refractions.
Here's the render with specular bounces raised to 12, in synch with the max ray trace depth. Also turned down the bounce light intensity a bit.
20 min 27.99 sec
The final one looks great! =)
And here is something I came up with recently. The glass is AllenArt's freebie, I just hid the bubbles, straw and ice cubes.
No caustics in 3DL renders.
So, first we have my glass shader with the path-length based absorption. You can see it doesn't work that well with the overlapping geometry that makes the fluid/glass interface work correctly otherwise ( http://adaptivesamples.com/2013/10/19/fluid-in-a-glass/ ).
Then... This is not a true nested dielectric scenario, but an approximation using trace subsets: the glass "sees" (reflects and refracts) everything, but the liquid does not "see" glass. As you can see, it solves the problem with path length based absorption approximation (the liquid "volume" is correctly shaded), at the cost of the glass not being reflected in the liquid. Just how exactly noticeable this is, well, it may become noticeable in some scenes...
It also shaves about a minute off the total render time.
We can see that the glass absoption is taken into account, too.
Now, about the actual refraction of the glass with liquid. Which one is (more) correct?
LuxRender gives a very similar result to the original 3DL render (plus with proper absorption and RT caustics).
Iray (with the original AllenArt materials) looks completely different.
Now... it's ridiculous but I don't have a glass... Could anyone replicate a similar scene IRL and take a photo?
Found this on wiktionary.
I'm assuming it's soda in a glass
, since that's what the original prop tries to emulate. I think neither renderer gives 'correct' result to the real life example.
Probably neither - the overlapping geometry is a hack either way. I found a couple more on a stock photo website:
http://www.123rf.com/photo_2548954_glass-of-ice-water-on-a-hot-summer-day.html
http://www.123rf.com/photo_1090864_a-glass-of-iced-water-standing-on-a-blue-and-white-tablecloth-ice-cubes-in-glass-and-condensed-water.html?term=icewater
Looking at those, I think the trace subset hack seems to sort of be more or less acceptable.
Some gamedev article claimed that for realtime, very simple distortions could pass for refraction because the human brain isn't really that good at processing refractive surfaces.
What looks kind of odd to me is that the refractions of the background are all different between the renders. I'm assuming you use the same IOR value. If that is so, then all three renderers treat the value differently.
Here's a shot I made with US2. I overdrive the glass specular to get a glint on the bottom of the glass. Works quite nicely as a fake caustic. Funny enough, the refraction of the grid looks close to the Luxrender shot.
Really frustrates me that none of the common shaders (Daz Original, US, US2) seems to manage to handle refraction + SSS
US2 handles them quite nicely. Just remember that SSS is tied to diffuse (though it doesn't have to enabled) and so, it's tied to opacity. As long as opacity is not zero, you can overdrive any diffuse tied parameters (diffuse, velvet, translucency and SSS). There's also the 2nd diffuse layer you can play with.
The second render has SSS applied to the liquid inside the bottle. SSS settings are the default (I only changed the scatter/absorption values and overdrive the SSS strength to 300% and backsatter to 10). Opacity is set to 10%. Third render is with SSS scale set to 0.1 - so less scattering overall. Fourth is with SSS scale to 10.
yes, but the fluid isn't cloudy.
I came across this when trying to make silty water -- as far as I can tell, you can't, not realistically.
SSS doesn't occlude with refraction.
Ubervolume doesn't work in a refracting substance.
... meh
I hesitate to make a comment because really quite clueless about the various shader options so ignore this comment if it makes no sense... but just in case this sparks a helpful idea, would it be possible to stack two materials within the same volume to partially fake the effect, by using the refraction from one shader's material and the cloudiness from the 2nd shader?
It's a great comment. I tried that. It mysteriously doesn't work. ;)
Well, at least not with the shaders I have.
If you put an ubervolume inside a refracting volume... it's invisible.
Easy. Just raise the opacity then use the 1st or 2nd diffuse. Alternatively, you can use translucency or SSS, or combine all of them.
It would help if you can show what real life materials you're targeting.
Here's some examples of mixing SSS with different opacity strength and opacity color. Remember that SSS is multiplied by diffuse, which is again multiplied by opacity. If you want to use SSS to 'block' the refraction, you need to raise the opacity strength. Use the opacity color to get more refined controls. I've kept the opacity color grey (ie 192,192,192 or 128,128,128) so all the color is coming from the SSS.
Is there something weird with ubersurface2? Because I can't get anything like any of those results.
Oh huh. It appears US2 IS broken, with SSS. That's fantastic, no wonder I was having so much trouble.
Could you give me some starting numbers? I tried your first example.
Set Ubersurface base on a cylinder.
Subsurface on, set to strength to 300% and backscatter boost to 10. Set opacity to 10%. Refraction on, 1.33
And it looks like a cylinder of water.
Different lights, progressive on/off, looks like water.
??