Sometimes, when solving the Void Rubik's Cube using the standard beginner method, it gets "stuck." What I mean by this is that the top edge of one face and the top edge of an adjacent face's colors are swapped.

Typically, I've just turned random faces until the cube is sufficiently messed up and then started over. However, there must be a better way - what is it?

  • 1
    $\begingroup$ The void cube is brilliant. Other puzzles derive from the Rubik's cube by adding to it—greater dimensions, latches, gears. To make the puzzle harder by taking away is delightfully mischievous. It's also a marvel it can be built. If you've ever taken a part a Rubik's cube, you know the mechanism sits in the centre. $\endgroup$ Dec 31, 2014 at 10:43
  • $\begingroup$ Of course you can emulate a void cube given a Rubik's cube and a marker pen. mrob.com/cube/cube-puzzles-5.html $\endgroup$ Dec 31, 2014 at 10:45
  • $\begingroup$ @ColonelPanic - Or just take a standard cube and ignore the centre cubies. $\endgroup$
    – h34
    Mar 14, 2015 at 9:48
  • $\begingroup$ @ColonelPanic: I read the linked wikipedia page and your comment but I am very unclear on why it is harder. Can you not use the exact same methods as you would use when solving a regular rubiks cube? The edge and corner pieces mean that the faces will all have the same relationship to each other so you could effectively just imagine the centre cubes if you wanted to make it identical? $\endgroup$
    – Chris
    Oct 1, 2017 at 23:17
  • 1
    $\begingroup$ @Chris Looking at a solved void cube, it's obvious what colours the missing centres would be, but handed a scrambled void cube, you can't tell. This frustrates ordinary methods with steps such as 'make a big plus' or 'solve a 2x2x2 subcube'. If the invisible centre of the plus happens to be wrong, you'll get stuck later on—pieces will be scrambled in an unexpected way your method can't handle. $\endgroup$ Oct 2, 2017 at 14:22

2 Answers 2


When you work with a void cube, you've de facto lost a degree of visibility. The algorithm to take you out of this position is M' U M' U' M U' M U2 M' - then you just solve the remaining edges as-is, and I'm going to attempt to briefly explain why this works.

You, as you know, can no longer see the centers on the void cube. However, it's important to bear in mind that they're actually still there, just hidden. When the 'virtual' centers are rotated incorrectly - that is to say, you're attempting to solve the cube around an invalid position - the pieces won't line up. In the case you're seeing, the centers along one axis have been rotated 90 degrees out of position.

To see for yourself why this might be, attempt to solve a regular Rubik's cube with one row of centers rotated 90 degrees. You'll find yourself in an equivalent position on the 3x3. (Note: a position which is mathematically equivalent to the one you are in is where two corners are swapped; if you end here, you're in literally the exact same position.)

This algorithm moves them back into a legal position, and, though it disrupts a couple edges in the process, actually allows you to solve it regularly again.

This can also be done intuitively. Simply do a middle slice along any face, then keep one corner fixed in your mind relative to the three virtual centers (i.e., keep it in the same place relative to them), and solve the cube again.

  • $\begingroup$ Thanks! Although, even when using the standard purple-blue-orange-green orientation (maybe different colors on different cubes), this still appears to happen. Why is that? $\endgroup$
    – Doorknob
    May 14, 2014 at 21:13
  • 2
    $\begingroup$ @Doorknob If your centers (even though you can't see them) are rotated in an invalid position, even if the color layout is correct, the puzzle won't solve properly. See my edit in the middle, starting with "to see for yourself why this might be" (grace period, apologies) $\endgroup$
    – user20
    May 14, 2014 at 21:15

I have long been frustrated by this, and found a very simple fix! As mentioned in other responses, this occurs since the centers are off by 90 degrees.
Instead of doing an algorithm, I merely do an M slice turn (turn the middle vertical 90 degrees) then solve the cross edges again around the new centers. This will work every time and requires no new algorithms!

  • $\begingroup$ Wow I just realised this is exactly what the standard M' U M' U' M U' M U2 M' algorithm does. First it does M', fixing the centers, then it puts in the back cross edge, and then the front cross edge. $\endgroup$ Jul 15, 2015 at 9:01
  • $\begingroup$ M is subjective, this may only work on 50% of cases $\endgroup$ Jul 29, 2016 at 18:50
  • $\begingroup$ @MonoThreaded I've done this many times and it has always worked for me. $\endgroup$
    – Kruga
    Oct 2, 2017 at 8:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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