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
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Sandy
Beach Towel model on HDRI scene....kinda.
This is kinda technical, so if anyone wants to know how I got topographic geometry from a pure single HDRI map I'll post a tutorial.
@charles, thats actually very nice. Thanks for sharing, would be interested in your process.
Create your HDRI scene with nothing in it.
Set Dome mode to FInite Box with ground.
Create a camera and position it up high looking straight down whre you can get as much of the area and details is you want. (It might be worth experimenting with different distortions and focals here but have not tried it yet.)
Render.
Move the render into a paint program and convert to greyscale and adjust contrast however you want for detail.
Take the greyscale into a 3d modeling program that allows the creation of a surface from a height map. I use Rhino 7, but there is some free online ones too I think. You will want the mesh poly count pretty high or it will look spikey.
Apply the original texture to the new detailed surface and align the UVs.
Save as an OBJ.
Import OBJ back into Daz. Align it to match the scene from the top down camera.
Turn ground off in render environment.
Apply any desired shaders and maps. You might also have to fiddle with the color settings to match the HDRI ground as your new ground no longer emits the HDRI information. (But MAYBE I need to fiddle with Infinity or some program that does support HDRI data.)
Create a new camera and position it in the scene and check your alignments and colors from ground level.
The towel isn't a towel, it's a double layerd Duvet with Fuzzy applied to it. The normal sheets and towels are simple planes that have no Y depth.
Align "towel" and simulate dforce, you may need to play with it for best results. Up collision probably helps. Might even want to subdivide the "towel" for greater detail.
I allow some poke throughs around the edges to simulate sand that his spilled over onto the towel.
That's it, you could go a bit further and...
Create and position a character on the towel, stick them a bit deep into it. Depth depending on how soft you want the sand.
Hide the towel.
Use mesh grabber and adjust down the parts of the ground mesh where the body intersects.
Hide the character and unhide the towel.
Redo dforce simulation.
Unhide the character.
--
As far as the sandy feet go, that's just 5 minutes in Substance Painter.
Import your character with UV tiling.
Create a fill layer that is just color and set it to solid black.
Create a paint layer and set it to Color/Normal/Rough/Height all on.
Find a good brush like Kyle's Splatter Brushes. (you might need to Invert the Height/Normal)
Paint your sand white. NO COLOR, just want a final color map that is just black and white.
Export your maps.
Create a geo shell for character, assign them all Uber, then opacity to 0 for all parts you don't want it applied to.
Find a shader like Sand or make one, rip a sand texture from goog and apply that to the effected body part surfaces.
Apply the correct body part black and white map to the body surface in Cutout Opacity, set that to 1.0 along with the normal height maps.
Done.
That tecnique can be used to make all kinds of stuff, from sweat, to blood, wounds, mud, dirt, scales on a character using the right shader and your own cutout. But why stop there? As you can see the vollyball has the same sand geoshell as the feet. You can apply scratches, roughness, streaks, spills, lint, condensation...etc. to normal objects as well.
Awesome, thanks for sharing your process.
just couldn't resist doing another b/w-fake-photo rendered portrait of Roxy in her kitchen.
-
Very lively!
P.s.
Maybe you would consider becoming PA? Heartbreaking to see how such beliavable characters are only customs
Thank you, I am glad you like her.
Small problem though. ROXY is an eternal work in Pogress projekt, which gets updated weekley by using pretty much everything I got from stores and freeby pages.
My usually workflow is to add parameters and materials, that makes her look more real and supports her designated character. Mostley also fixed and enhanced with custom ZBrush fixes. After a period of testing, those parameters get baked into her and new stuff got added to her.
So, because of legal issues you might understand, why she impossible to be released, no matter what.
But thank you for asking.
Ah, i thought you have several different characters in works.
Well, keep us updated about new Roxy adventures, then
This is exactly how I set up my SSS, I set the translucency to 100/1.0 which completely takes over the diffuse, and simply use the diffuse map as the translucency map, it's much better for use with the specular rendering mode!
Also, I wish the poser/Daz community would quit obsessing over the backscattered ears FFS, if you're using the SSS properly you won't need to worry about those bloody glowing ears as there are much better ways of telling if you're using SSS from looking at the front of the character and not obsessing over the glowing ears!
It's why most of the bad SSS settings have the character's nose glowing unrealistically because the glowing ear crowd always tune the settings for those glowing ears!
Ironically, glowing ears are pretty much how I set up my skin settings though. Below is basically the lighting I set my sss up in. Heavily backlit but with a small light from the side as well to make sure the nose isn't jello. I generally find that if I set up with this light I dont have to worry how it will behave in more neutral lighting, but If I set up in neutral lighting its much harder to judge if the sss is right and bad settings can look okay in neutral lighting, and you don't notice that the neck is jello
I just don't aim for glowing ears in every situation, though. In reality it only happens with strong backlighting, so If you try to get glowing ears with neutral lighting, yeah bad settings are guaranteed. But I would absolutely describe myself as "obsessed with backscattered ears" If the ears don't glow when properly backlit your sss settings aren't right and will be off even in neutral lighting. Backsattered ears are an integral part of my workflow
for ref my skin sttings vs a default 8.1 character
default
my preset
The nose glows less, but I would argue the ears are a bit brighter
(the new defaults are definitely the best defaults yet IMO)
For me, I measure it with cast shadows and skin waxiness, here are my SSS settings, I based it on this method, carrying it over to the PBR settings... Here is G8.1 with my sss settings.
I just do a decant amount of dramatic lighting
lots of this
so I don't feel comfortable unless I know my settings work in the most extreme conditions, if I worked with your lighting I wouldn't feel confident that I could shine the full light of a thousand suns at the side of the face and not have the nose glow. very much just a workflow preference though.
unrelated but your settings remind me of my biggest pet peeve with the PBR Skin Shader.
Does it actually effect anything? Probably not. But why? Why aren't both displayed as linear values? Why daz?
I too like using varying light sources, my favorite is a cloudy sky HDRI as well as a normal sunny HDRI and the sun&sky shader, as far as why doesn't Daz use the linear values... Daz is gonna Daz!
But yeah, I found that those settings have completely cured the seam issues when using Specular rendering ever since I followed this guide...
The thing is DS (or Iray at least) does use linear values and every singly other color in the shader displays the linear values. as shown circled in the transmitted color. Its purely a visual thing that the sss color is displayed differently. Its ultimately just an annoyance when setting the color (Although not for me because I went into the shader in notepad++ to change how it was displayed)
And yes I am fully in the church of the .99/.99/.99 transmitted color
OMG, yes, although what I'd really love is to be able to work in HSV, but consistency for settings across everything would be wonderful. No more 0-1 and 0-100%. One or the other.
This looks pretty good. The necklace and shoulder strap hovering are giving it away though :)
I think this is the problem with someone with super smooth, mathematically symmetric features, and clean textures, is how to insert detail into it to make look realistic.
I'm a little skeptical of the "too perfect for realism" school of thought. We're so used to manipulated images of people, where irregularities and "blemishes" are removed, are we really seeing 3D as fake because of that? Obviously, we're seeing something, and that something is probably subtle, so it could be there's a certain kind of 3D perfection (or simplification) that reliably clues our brain to the fakery.
This picture is obviously Photoshopped to the point where we don't see it as representing the woman. But I, at least, don't think it's a 3D render. It looks like a heavily manipulated photo. Masterstroke's image is better than this and has more skin detail, but looks 3D. My first thought is that there are subtle facial geometry issues at play, but given that since the dawn of photography, retouchers have been "fixing" facial features, I'm not convinced. Even with apps that cartoonize photos with huge eyes, it usually looks different than a 3D render given the same treatment.
To be honest, i'm baffled by it.
Well all that they simplified in that photoshopped image was some of her skin texture. There is literally a huge amount of high frequency detail left in that image. look at the insane amount of detail in her hair, eyes, skin. Particularly, Look at all the detail in her eye ball, eye makup, lacrimal gland. None of that was in Masterstroke's image. To say nothing of transitions between different "textures" like lips and skin. And look at her nose "morph" and mouth "morph". Those are complex shapes. Her jaw line is complex. Her eyebrows are complex.The peach fuzz on her faze and microdetails soften the edges, which was one of my critiques of masterstroke's image (sharpness).
Photoshop into that image some daz eyes and hair and see how realistic it looks.
Hmm… I'm not buying it. Looked at small enough that you can't see the details, the horribly Photoshopped image still looks like a photo of a person. And as far as the "ugly" Daz eyes, morphs, etc., the ArtStation render of Vicky 8 (8.1?) by the Daz Artist who sculpted it is very convincing. So it's not Daz shapes, at least not alone.
Maybe it's just something that artists who master this have absorbed intuitively via study, a particularly focused artistic eye, or something else that I'm missing, much in the way people who a lot of listening practice can pick up nuances in tone that most people can't, or a trained coffee taster catches elements that are part of what the average coffee drinker asigns to "coffee taste" or "sour" and "bitter."
Yes, over here, I would love a text-based tutorial!
Literally every single person ive seen who makes photoreal renders does so on the basis of detail / realistic shaders. Zooming out so you cant see the detail doesnt mean it's not informing how the pixels smear together during the resize algorithm/ artistic filter applied in post.
The work she did on Vicky 8.1 is good. But kind of something uncanny about it? Her eyes are incredible, but perhaps too unrealistic? I think the hair is doing a lot of the heavy lifting in that render. We dont have any good hair in Daz equivalent to a high quality Ornatrix groom etc.
Mostly for my own understanding, here is Masterstroke's render, Hyseon Oh's, and the bad photoshoped photo reduced in size so details are lost, and toned to similar contrast levels.
At this scale, it does appear both skin shader and geometry matter. The photoshopped photo looks the least like a real person to me with the skin smoothed over so heavily. Your point about eyes makes more sense when the original Oh render is compared to this, Blue, becuase this looks more convincing when I can't make them out well. Two things strike me about Masterstroke's character—one is this pose (or lack there of) makes it look less real because nobody holds their shoulders like that normally, and the rounder, softer face implies more body fat, which isn't there in the neck, so that reads as off. But put a shirt on her and throw that in an old black-and-white yearbook or other gallery of small black and white studio portraits and don't think many people would spot the fake.
I just love this discussion and the wonderful examples posted. I am taking notes! My take on this, apart from exploring and applying special shaders, normal maps, subsurface scattering, light, shadows and and other techniques, is that convincing photorealism in people and portraits is best achieved when we allow a believable "person" to emerge from the work, though realistic expressions, imperfections in face and body shapes, mannerisms and posing. It's all subjective, of course, but that's when it all comes together for me. In this case, this girl, that I conjured up, would be the exact kind of girl I would have totally crushed on in high school or college. I see a person, not a 3D character. That's when the work "comes alive" for me, as we say.
Cheer all.
At that scale, third one still looks most real to me because the pose is very natural, morph is natural, albeit super model-y, and most of all the level of detail in the hair simply cannot be matched.
If we look at a photo taken on a potato camera, we still see it as photoreal because what the camera captured was real life which exists at like 30,490,305,343 pixels and is the culmination of very small subatomic detail. Yes the camera will not capture every detail, but what it captures is still because of that detail. So imo shouldnt your render immitate that detail even if you dont think you will notice it? Anyone who tells you that you that you dont need to turn on normal maps (/HD morph) if you're zoomed out or use high-geometry hair is probably lying to you.
Ha. I should have cropped out all of the hair, because none of the Daz hair I'm aware of get super close to photoreal. But I'd say the third one has all the details removed. I wonder if there's a bias that comes from persuing super detailed renders that leads your eye to look for things you might not otherwise. I just learned that when we learn to read, the reading-related shape recognition highjacks the parts of our left brain that go into facial recognition, so that literate people only use their right brain for facial recognition (and are worse at it as a result). It could be as you train your brain to see details, you start perceiving things differently than people who don't train themselves. Kind of like how an X-ray or ultasound professional can spot things non-pros can't.
I'm going to quize some friends who don't do 3D art and see if they think which, if any, of the images are 3D.
Awhile ago I posted mei lin 8. I still think her textures are amazing. Here is a raw render in blender, the only thing i changed for her texture was the brightness and saturation.
But what i really think is photoreal is AI rendering. Forget trying to render photoreal . someday AI will do it for you, instantly.