How are 10k images rendered for NFT?

Question

3D rendering

Greetings. I am writing a question with a deep awareness that this will be another case with 10k of unnecessary images, but I am still bursting with curiosity how the process of creating so many images is arranged in a 3D style. And in general, such a skill will not be superfluous.
With the drawn two-dimensional images, everything is clear how you can give out 10k images through a randomizer generator, but this is not interesting, I want to learn how to do this in a blender or blue.

Of course, you can pull out individual parts of the character as png with alpha, and then randomize it through the code, but a lot of problems will come out with incorrect overlapping of objects and shadows. In general, how do trideshniki do it? An example is below.

Answer 1

You can, of course, draw out individual parts of the character as png with alpha, and then randomize it through the code

That's exactly what they do. If you have a problem displaying 3D with alpha norms ... well, people. These are the basics.
There are no problems with shadows, if the hands are not from the ass.

problems with incorrect overlapping of objects and shadows

To do this, everything is done in groups. First the bottom layer, then the top layer on top of it, then the accessories on top of it
. That's what they definitely don't do is 10k pictures.