WCA Documents: Updated January 1, 2013

[Stefan]

Since the gap could move to any place on the layer from which the piece popped, it can be placed anywhere in that layer. Where as on a 3x3, the gap would be stuck between 2 edges or corners, and since they can't interchange freely, the gap remains in the same relative place. Since it can be placed anywhere, the worst case would be a DNF, even though the orientation is fixed.

Why should you move the gap? That doesn't even seem permitted.

What about a popped 4x4 edge, edge+center pair or edge+center+center triple (in a straight line)?

 

[ThomasJE]

Why should you move the gap? That doesn't even seem permitted.

What about a popped 4x4 edge, edge+center pair or edge+center+center triple (in a straight line)?

Normally, 4x4 edges only go in one way, and centers would just go in either of the two gaps; the cube would still be solved afterwards. That's what I think; but whether it is right I don't know.

 

[clement]

As for the center caps, it doesn't make sense now that I think about it. I could solve the cube having only 2 center caps on (as opposed to a void cube, which can have a parity problem) every time. Of course it's not the fastest way, but it can be unambiguously solved. If 2 center caps fall when the cube hits the table as I stop the timer, and the rest of the cube is solved, I don't see the point in making it a DNF.

The problem here is that you require that the judge has enough knowledge to know that it is unambiguously solved. If you write the rule like you suggest, then a lot of judges would not be able to judge if it is solved or not.
If this case is too simple for you, consider the case where you remove 1 corner cap from every corner, it is still unambiguously solved.

 

[PatrickJameson]

The problem here is that you require that the judge has enough knowledge to know that it is unambiguously solved. If you write the rule like you suggest, then a lot of judges would not be able to judge if it is solved or not.
If this case is too simple for you, consider the case where you remove 1 corner cap from every corner, it is still unambiguously solved.

It is the competitor's responsibility to know the rules. If there is ever a question, the competitor can still ask for the delegate to make the judgement call.

 

[DrKorbin]

AFAIK, in all modern 4x4x4's an edge physically cannot be placed on its place with wrong orientation, so a popped edge would lead to no penalties.
But if tomorrow someone will create a 4x4x4 that has "symmetric" edges that can be placed on their places twisted, then a popped edge on this particular cube will lead to DNF, right?

Normally, 4x4 edges only go in one way, and centers would just go in either of the two gaps; the cube would still be solved afterwards. That's what I think; but whether it is right I don't know.

I think Stefan meant that after an edge+center popped (for example, white center and white-green edge), you can move green center adjacent to a gap on the white layer, just like you move pieces and gaps in popped square-1.

The problem here is that you require that the judge has enough knowledge to know that it is unambiguously solved.

Not the judge, but the delegate. If a regular judge is in doubt, he calls a delegate.

 

[MaeLSTRoM]

Why should you move the gap? That doesn't even seem permitted.

What about a popped 4x4 edge, edge+center pair or edge+center+center triple (in a straight line)?

It wouldn't be permitted, but the competitor might have it somewhere else in the layer, pop it intentionally (might not be able to tell as a judge if this happened) and then move the gap around so the edge would be solved, while making it look like an accident. Since there would be no way to guarantee where the piece came from, it should be a DNF, at least that's how I think it should be ruled as.

As for the 4x4. single edge would be solved, 2 edges would be DNF since could be OLL parity. Also, any amount of centres from the same face is fine. Also fine is lots of centres, as long as all of each individual type of centre is the same colour (e.g. red xcentres, green +centres, blue left obliques, white right obliques would be fine.)

 

[Julian]

The problem here is that you require that the judge has enough knowledge to know that it is unambiguously solved. If you write the rule like you suggest, then a lot of judges would not be able to judge if it is solved or not.
If this case is too simple for you, consider the case where you remove 1 corner cap from every corner, it is still unambiguously solved.

Perhaps you should be given the chance to prove to the judge (or delegate) that the cube is indeed unambiguously solved?

 

[ThomasJE]

I think Stefan meant that after an edge+center popped (for example, white center and white-green edge), you can move green center adjacent to a gap on the white layer, just like you move pieces and gaps in popped square-1.

Ah; okay.

 

[Stefan]

I think Stefan meant that after an edge+center popped (for example, white center and white-green edge), you can move green center adjacent to a gap on the white layer, just like you move pieces and gaps in popped square-1.

Yep, that's what I meant.

As for the 4x4. single edge would be solved, 2 edges would be DNF since could be OLL parity. Also, any amount of centres from the same face is fine. Also fine is lots of centres, as long as all of each individual type of centre is the same colour (e.g. red xcentres, green +centres, blue left obliques, white right obliques would be fine.)

Not sure whether you avoided the interesting cases (edge+center(s)) or answered inconsistently (to your Square-1 argument).

 

[MaeLSTRoM]

Yep, that's what I meant.



Not sure whether you avoided the interesting cases (edge+center(s)) or answered inconsistently (to your Square-1 argument).

Ah, no. I misread the post I replied to. However on a 4x4, the centres wouldn't slide because the other outer pieces hold them in place, since the 4x4's centres always point towards the centre of the side, they would have to twist past another piece to move, so if only the edge+centre came out, it wouldn't be a problem since motion is impossible, and should be no penalty. Same goes for the 3 piece line, although I think on some mechs that are looser than others, there might be cases where centres could move, so I'm not so sure on that one.

 

[DrKorbin]

Ah, no. I misread the post I replied to. However on a 4x4, the centres wouldn't slide because the other outer pieces hold them in place, since the 4x4's centres always point towards the centre of the side, they would have to twist past another piece to move, so if only the edge+centre came out, it wouldn't be a problem since motion is impossible, and should be no penalty. Same goes for the 3 piece line, although I think on some mechs that are looser than others, there might be cases where centres could move, so I'm not so sure on that one.

Try LanLan. Some centers can slide.

 

[Kian]

What? Why? As long as two adjacent center caps are on, the rest are irrelevant because the cube is unambiguously solved.

"5b5) If parts of the puzzle are still defective or not fully placed at the end of the attempt, the result is recorded as the worst state obtainable by repairing the puzzle "

This includes states that wouldn't normally come to be on a puzzle. One center cap is fine, two is a DNF. This is new.

 

[tim]

So, "repairing" in 5b5 basically means "putting stuff back to where it fits"?

 

[MaeLSTRoM]

So, "repairing" in 5b5 basically means "putting stuff back to where it fits"?

I think this is probably the most straight-forward way of putting it. If you can put stuff back in such a way that the puzzle isn't solved, then its the relevant penalty (+2 for pyra tips, DNF for most other things, etc.). THe only other case is when no matter how the pieces are replaced, the puzzle is solved, and so there is no penalty for this case.

 

[DrKorbin]

I think this is probably the most straight-forward way of putting it. If you can put stuff back in such a way that the puzzle isn't solved, then its the relevant penalty (+2 for pyra tips, DNF for most other things, etc.). THe only other case is when no matter how the pieces are replaced, the puzzle is solved, and so there is no penalty for this case.

But this will depend on the model of the puzzle. You have already given an example when red x-centers and blue oblique centers pop. What if my cube allows to put oblique center in the gap for x-center? Well, the cube will look weird and won't be cubeshape anymore, but still it is "putting stuff back to where it fits".

 

[uberCuber]

"5b5) If parts of the puzzle are still defective or not fully placed at the end of the attempt, the result is recorded as the worst state obtainable by repairing the puzzle "

This includes states that wouldn't normally come to be on a puzzle. One center cap is fine, two is a DNF. This is new.

With that particular post I actually didn't intend to question what the regulations say; I was questioning the logic behind it. As some people have said since I wrote that post, it makes little sense to call a DNF with only two center caps fallen off, because any slightly knowledgeable cuber should be able to tell that the cube is still unambiguously solved. Like someone said, I could take four center caps off my cube (such that the remaining two are adjacent) and just throw them away completely, and I would still be able to solve the cube without problems, nothing would be changed at all (except for slightly more difficult recognition or whatever :s)


That aside, I do also find the use of the word "repair" in these regulations very questionable. I have absolutely no problem with the idea of DNFing a cube that has a twisted corner or a popped edge when the timer is stopped, but I don't like the way the issue is being handled here. To "repair" something is to fix it, it's the definition of the word. If a single corner got twisted clockwise, twisting it clockwise again doesn't fix anything; calling it a "repair" is nonsensical. By the actual definition of the word repair, the only possible repair in this case is to twist the corner counterclockwise - to fix it.

 

[Stefan]

As for the 4x4. single edge would be solved

Why not +2? You can move the other pieces around just like with Square-1, ending up with a turn away from solved. Or even two turns, so shouldn't it be DNF?

 

[Dene]

Why not +2? You can move the other pieces around just like with Square-1, ending up with a turn away from solved. Or even two turns, so shouldn't it be DNF?

I don't want to jump in on this discussion, I just want to point out there is a difference. On a Square-1 you'd move the pieces within the layer. On a 4x4 you'd move the pieces by moving the layer. Very different.

But please continue . I too am interested in the results of this discussion.

 

[DrKorbin]

I don't want to jump in on this discussion, I just want to point out there is a difference. On a Square-1 you'd move the pieces within the layer. On a 4x4 you'd move the pieces by moving the layer. Very different.

I've almost succeeded in turning one internal layer in LanLan 4x4 with popped edge+center by moving the pieces separatery. Unfortunately, the cube crumbled to pieces -_-
Besides, regs say nothing about "turning the pieces within the layer" and "turning the pieces by moving the layer".

 

[Stefan]

I've almost succeeded in turning one internal layer in LanLan 4x4 with popped edge+center by moving the pieces separatery. Unfortunately, the cube crumbled to pieces -_-
Besides, regs say nothing about "turning the pieces within the layer" and "turning the pieces by moving the layer".

Darn, I was about to say the regs don't distinguish that, but you were faster

I successfully turned (far enough for DNF) an inner layer of a Rubik's 4x4 with just a missing edge by moving pieces separately.