[gamblor956]

the simplest is just mixing filaments, like one mixes paint.

That is actually the hardest way to do it, because that's not at all how 3d printing works. 3d printers take strands of plastic (aka "filament"), soften them up to being melty but not melted, and then "extrude" them, like cake frosting onto a surface. As with cake frosting, in order to mix colors, you have to do so before the extrusion step, so you would have to make your own (filament), and the machinery to do so is not cheap.

The thing about first order thinking is that it is very rarely useful, because the actual experts in the field have almost certainly thought of all the things that first order thinkers come up with, and deemed those ideas unworkable for various reasons.

[MayeulC]

Hence why I was asking the question, thank you for answering, I did not consider the high viscosity of the filament when I wrote this. As you say, it can be worked around, but that would require extra complexity/cost.

[kayson]

I wonder if you would get good enough results by just extruding two separate filaments simultaneously. Sure, they won't fully mix, but with thin enough layers, you'd benefit from the same visual processing that makes alternating layers look like a solid color...

[jononor]

Dual color filaments exist, and they do not mix at all... It gives the objects a nice transition when rotated. But indicates that color mixing in the nozzle is probably pretty difficult?

[zahlman]

> The thing about first order thinking is that it is very rarely useful, because the actual experts in the field have almost certainly thought of all the things that first order thinkers come up with, and deemed those ideas unworkable for various reasons.

Sure, but a useful article would focus on explaining the consideration and rejection of those obvious ideas, and what actually had to be done to implement something similar — rather than focusing on the even-more-obvious background material (how 2d printers have worked for decades) motivating the obvious ideas.

[gamblor956]

That is a very different article from the linked post, which was about introducing the cool new tool someone else developed into their own program.

If you want a discussion about the other ideas and why this specific method was chosen, go to the original tool where the creator discusses those things.