Is there an infinite analog to the Rubik's Cube? What does its solution-algorithm look like? For illustration, consider the Rubik's cube with infinite tiles to a side, on all sides, with sides of finite length.

  • 1
    $\begingroup$ Problems might just be slices at real valued intervals through the cube rotated iteratively until a confused state is reached and then solving it. $\endgroup$
    – Dohleman
    Jan 8, 2019 at 1:58
  • 3
    $\begingroup$ You might want to look up Thompson's groups $F$, $T$ and $V$, and also the Brin-Thompson groups $nV$. They are not exactly what you are looking for, but I think they might be close. Certainly, they are "slices" of the interval $[0, 1)$, and the groups $nV$ are slices of the $n$-cube $[0, 1)^n$. They also have rather pleasing graphical representations. $\endgroup$
    – user1729
    Jan 8, 2019 at 11:55
  • 2
    $\begingroup$ You have to say what you mean. For example, it is the cube $[-1,1]^3$, but rotations can be made fixing any subset of one of the axes: rotate the planes corresponding to that subset, fix the rest of the planes $\endgroup$
    – GEdgar
    Jan 18, 2019 at 11:42
  • 2
    $\begingroup$ I wonder how an infinite rubik's cube can have finite sides? $\endgroup$ Jan 21, 2019 at 3:14

1 Answer 1


I made up two variants of infinite Rubik cube. One uses the space $\bigl(\Bbb{Q}\cap[-1;1]\bigr)^3$, which has countably many tiles. The other one uses the space $[-1;1]^3$, which has uncountably many tiles. In both cases, each face has its own colour and a twist is performed by rotating a slice which is orthogonal to any of the axes. The cube is solved if and only if all tiles on each face have the same colour.

If the Rubik cube is scrambled by finitely many twists, the cube can be discretised in such way that it is symetrical and each great tile has only one colour. After that, it can be solved as a Rubik cube with finitely many tiles.

The problems starts when the Rubik cube can be scrambled by infinitely many twists. Each scramble can be represented as a bijective function from $R$ to $R$ where $R$ is either $\bigl(\Bbb{Q}\cap[-1;1]\bigr)^3$ or $[-1;1]^3$. The task is to decompose such function to an infinite sequence of twists. There are uncountably many scrambles, so it might not be even algoritmically solvable.

  • $\begingroup$ Given the ambiguity and modesty of the question statement, I find this answer acceptable. Thank you. $\endgroup$
    – Dohleman
    May 20, 2019 at 9:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.