How did this happen?

[FrankLZ]

Ok, so I took off all the center caps for one of my DIYs so I could mess around with it and adjust the tension but something weird happened.

I solved it and got this:

y2:

Down face = white

Then scrambled it and solved it again and got a perfectly solved cube. I didn't disassemble the cube and move the pieces around.

How did this happen?

PS - I have also gotten the case where two opposite edges are the only pieces switched.

 

[Ellis]

Because you still have your center caps off.

Edit: See void cube

 

[Waynilein]

M' U2 M U2 M' F' L' F (M' U)x4 F' L F

If you ignore the centers, a slice turn will cause odd permutations, in the sense that 2 pieces can end up being switched.

 

[FrankLZ]

Shoot, I was hoping to stump people for at least a little longer than 15 minutes.

I totally expected that as long as you solved the cube with the colors in the same relative position then the specific color of the centers wouldn't matter.

 

[blgentry]

Said a different way, try solving a cube (with center caps) but solve the white cubies around the blue center cap. Then keep solving based on those cubies. You'll end up with a condition like you showed above, except that the center caps will be in place, making the cube look funny.

I didn't know this before you posted, and just guessed it was true, and verified by solving a cube as I described above. Neat.

Brian.

 

[Waynilein]

Try solving a cube (with center caps) but solve the white cubies around the blue center cap. Then keep solving based on those cubies. You'll end up with a condition like you showed above.

Wrong. Hold the cube with green in front and white on top and do M S' M' S and you'll end up with white cubies around the blue center, otherwise fully solved.

 

[blgentry]

^ Hold on there professor. Read what I wrote again. I said that if you solve as I describe, you'd end up with a condition like the OP showed in his pictures. Namely with two edge pieces out of place, centers wrong, but otherwise normal.

Give it a try and you'll see.
EDIT: It seems that there's a key I left out: After solving the cross around blue, align at least one edge with a center cap. Then you're guaranteed to have the "parity condition". Otherwise, you may end up with all wrong centers and no "parity".

What you said is true, but it does not describe the condition that the OP showed, or that I described how to reproduce.

Brian.

 

[Ton]

M' U2 M U2 M' F' L' F (M' U)x4 F' L F

If you ignore the centers, a slice turn will cause odd permutations, in the sense that 2 pieces can end up being switched.

The M move will change for edges from position -you have no center-, this is an odd permutations

So if you do a M move

And from this position you can solve as usual

Some principle is there for a 4x4 is you move only a middle layer, solve the center and you will change 4 edges; hence a odd permutations

 

[TomZ]

Ton, that's not quite the truth. I think it could hapeen but the 4x4 permutation 'parity' is in fact an even parity situation and does not depend on the permutation of any other kind of piece. Try doing the PLL parity fix on a 5x5.

 

[Offblast!]

I have had it happen to me once that I inadvertantly rotated one corner piece in it's place making the cube unsolvable.

When it came to the PLL my brain started melting but I managed to figure out what was going on before things went bad.