Generalized shape inversion applied to the Mandelbox

[CozyG]

@ CozyG
I  posted a 3D frag of the simplest cuboid minR2 version  and it is identical to my Mandelbulber V1.21 OpenCL version (except rotated 90 degrees.)   Have not tested it much.  I am sidetracked on to  trying out some other things.

Very cool! Thanks for trying out shape inversions in 3D, and for making a frag too so I could more easily follow the code. I am hoping to contribute to 3D development soon. But like you I've been sidetracked I'm exploring some further ideas in 2D while they're fresh in my mind.

[Alef]

I just did something simmilar with inner circle. But based on superelipse Lp spaces and superformula. And alsou like "    
2D Mandelbox: Separating radius of inversion from max radius "
I will share soon. I did that not looking at these threads.

[Alef]

Here what I did. I hadn't spend mutch time on render so it is not gorgeous.
Alsou formula hawe lots of sssloowww functions. So I made "smart formula" with conditionals so it can awoyd some of these.

Maybe it is final version maybe it is not. It depends on do I found any better mean values. Mandelbulb3D works. I did only sphere inversion not full spherefold. Sphere inversion seems to be the most important part and alsou the less noisy.

Power 1 mixed with power 2 and power 3 generates some dust but it don't fully destroys DE.