Rendering "Minatures"--- scale up vs scale down?
Scavenger
Posts: 2,674
I'm doing a scene where the focus characters are supposed to be "minatures"...figurines..etc..
IS THE BETTER WAY TO GO, to scale Them down, and place them in a regular sized setting (table, stuff around them). Or to keep them regular sized, and scale up the background stuff?
Does it matter? Will one way give me better performance from the program?
Will one way give me better detail on the figures?
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
© 2026 Daz Productions Inc. All Rights Reserved.
Comments
If you're using a limited depth of field to accentuate the toy-like effect, you're better off scaling the mini-characters down rather than scaling everything else up. Actually, what I'd do is set up all the figures that will be miniaturized by themselves, parent them to an invisible plane primitive, then shrink them all down together by scaling the primitive up and down.
I haven't really decided on DOF yet..likely will to some level to blur the background stuff beyond the table, but kind want the figures as real as possible...they're figures, but thery'e the reality, type of thing..
Thanks for the idea of the plane.
Depth of Field can make anything look like miniatures, regardless of the scale to background (at least IMO), as it can make them look like their are in a dollhouse even if the room they are in is in the same scale as the figure. But scaling up the background and DOF together, esepecially if you put them on a tabletop in a room or something, would really sell them as being miniature I think.
A couple of other points. If you scale everything else up, panning and zooming to set up the cameras will get rather tedious. If you shrink, then on the small figures, watch out for shadow bias. You will need to adjust it, or the small figures may be missing some or all of their shadows. I had this problem a while back when I had a 50:1 ratio of full size to miniature figures.
One advantage of keeping the figures normal-sized and scaling everything else up is that the figures probably have the most complicated shaders and they are going to be the focus of attention more than the environment. For a simple shader maybe only the bump/displacement/normal strengths need adjustment, but a complicated shader may need a lot more tweaking to work the way you want at drastically different size.
Well, realistically, for a lot of angles the most practical option would be to do two renders, with the mini-items on a clear background with a shadowcatcher, and to put everything together in photoshop. Something to keep in mind is that light also scales proportionately, and the bigger something is, the more natural fill light and haze will occur to lighten the shadows. Pitching that inversely, extremely small items will not only show the effects of light with a far greater contrast ratio for the shadows, but would also become more translucent due to the decreased density. An extra pass to increase the subsurface effects would be in order if you were trying to be 100% realistic.
Right now I'm going with the "big stuff" scaled up, but the small stuff scaled down a bit...like a 300%/75% split
I'm running a higer quality render over nite, so I'll see how the shaders come out.
I set up the two sections seperatley...made the "big stuff" scene, set the camera, then all of it, camera included, was parented to a plane for scaling, so the camera doesn't need reseting. Then merged in the minatures who had been previously set up and grouped so they could then be scaled too.
Cybersox: Not sure I understand what you mean re: the light scaling...guess I'll see how the better render comes out.
I also don't know what a transparent shadow catcher is...how is that a thing? possible?
in DAZ studio it uses a shadermixer brick, carrara and Poser have shadowcatch included
unfortunately the recipe for DS is on the old forum, I made one but unsure about shareability
A shadow catcher is a plane or surface that which is invisible in a render except for the areas in which a shadow is being cast on it. It's especially useful for combining 3D rendered images with photo backgrounds. There are a couple of free ones out there on various sites, but DAZ also has one for sale here: http://www.daz3d.com/pwcatch
Light scaling is simply working with the fact that as an object gets smaller or bigger, the way that light passes and reflects through its outer surface changes. For example, if an adult holds their hand up in front of a bright light source, the skin and flesh on the shadow side take on a redish hue as some of the light is actually going all the way through the hand and illuminating it from within. The bigger the hand, the less pronounced the effect it is, so if a baby's hand were held up next to the adult's in the same light, it's going to look even redder and even the bones inside may be show up as darker shadows. Conversely, if a 20 ft tall giant's hand were held next to the adults hand in front of the same bright light source, the flesh of that hand would be much, much thicker and much less light would pass through so that you would only see reddish effect at the very edges if the tips of the fingers, while a 40 ft tall giant's hand would have so little translucency around the edges that it would start to look dead. Like depth of field, it's one of those visual cues that our brains use in order to determine the actual size of an object. DS, Poser and other 3D programs give you the ability to manipulate the same kind of visual effects using subsurface shaders, but a lot of people tend to overuse them and make their characters look less real by making them appear overly translucent.
By the way, in regards to my earlier comments about DOF, you might want to look at the wikipedia entry on miniature faking to see how depth of field can be used most effectively in making even very large objects look very small: http://en.wikipedia.org/wiki/Miniature_faking
One thing to remember, no matter what you do, is displacement maps.
Figures scaled down need their displacement map settings likewise shrunk.
Figures scaled up... well, yeah.
If a dragon is textured to look like its plates stick out two inches on its twenty foot long full size via displacement, then shrinking it down to six inches and the plates become spikes 1/3 as long as the figure is.