The preview image, as you know, is a rubix cube, a 3x3x3 puzzle.
What I (with the help of my supercomp) have created, is a Five-dimensional rubix cube, or 3x3x3x3x3
We label the puzzles like this because they are a d-dimensional cube broken into 3d smaller pieces or "cubies" of the same dimension. For example, the 3D cube has 33 or 27 total 3-dimensional cubies.
Each of the d-dimensional cubies could be considered to have its faces covered by stickers of one smaller (d-1) dimension. But each cubie also only exposes a subset of its stickers to the "outside", meaning these are the stickers you could see if you lived and operated in d dimensions. We can use the number of exposed stickers as a classification of cubie types. For the 3D case, the 27 cubies are broken into 4 types, those that expose 0 stickers, 1 sticker ("centers"), 2 stickers ("edges"), or 3 stickers ("corners"). Each sticker on a given cubie has its own color, so we could also call these 1-colored, 2-colored, etc. pieces.
In general, a d-dimensional cube will have d+1 of these types, those that expose 0,1,...,d different colored stickers. By starting with the number of pieces in the 3D, 4D, and 5D puzzle versions, one could perhaps extrapolate a formula for the number of each of these types of pieces for any d-dimensional Rubik's Cube (It was easier for me to arrive at this by looking at vector representations of sticker coordinates on the 5D cube). In any case, the formula for the number of a given type is:
2s . dCs
where,
d is the number of dimensions,
s is the number of stickers on a given type, and
dCs means the number of combinations of d things taken s at a time, equal to
d!
s!(d − s)!
For example, to get the number of edges (2-colored pieces) on a 33 cube, we have s=2 and d=3 so,
number of edges = 22 . 3C2 = 4.( 3!/(2!(3-2)!) ) = 4.3 = 12
Another cool observation is that with this formula, we can go backwards and easily find the proper analogies for 2-dimensional, 1-dimension, and 0-dimensional Rubik's cubes. These aren't very exciting puzzles because none of them can actually be scrambled, but thinking about their cubie types is a little interesting.
Using the above, you can then also find a piece counting formula for "Rubik's Revenge" (4 divisions per side instead of 3), "Professor Cube" (5 divisions per side), etc. versions of any d dimensional Rubik's Cube. If n is the number of divisions per side, the puzzle has the form nd and everything gets multiplied by a factor of (n-2)(d-s). The full formula is:
2s . dCs . (n-2)(d-s)

if this makes any sense at all.
Comments
pbjplatypus Says:
Holy crap! This is the first time saying this to anyone, but you have more time on my hands than I do.
Become So Numb Says:
Damn. Way to go Trev!
Sad thing: This makes sense to me. And I'm not super genius like you. o.O And I don't understand it, but it makes sense
R aKu kY o Says:
to be honest... it looks on the lines of hypercubes.
Neji Luver101 Says:
That's really cool. :3 Awesome.
koiblab Says:
this hurts my head...
Ixion Industries Says:
I'm afraid that, being only wise, not intelligent, this makes no sense to me.. but it does work as a good halluciogen..
k Y r O Says:
Oh dear god my brain!
You, have WAY too much time on your hands trev.
*Stumbling around dizzily*
Tengu Chidori Says:
i looked at this with 3d glasses. BLEW MY FUCKING MIND
Mangacritic001 Says:
sick man...
jess vs people Says:
just blew. my. mind.