Do we mind AUFs in PLL time attack?

[Godmil]

Ok, I'm more tolerant of the other view now. I always saw it in the way that the algs swapped certain pieces and any U moves at the end could be cancelled out by an AUF depending on the case. I guess it's perfectly valid to view a PLL as a sequence that solves a particular LL pattern.

 

[Forte]

Yay

 

[Jaycee]

Ok, I'm more tolerant of the other view now. I always saw it in the way that the algs swapped certain pieces and any U moves at the end could be cancelled out by an AUF depending on the case. I guess it's perfectly valid to view a PLL as a sequence that solves a particular LL pattern.

What exactly changed your mind?

 

[cubernya]

While we're at it, I'd like to point out something about the usual PLL arrow diagrams that I find inconsistent. Most cases minimize arrows, but not the G-perms. They could be done so that two corners are solved and two need to be swapped. Then we'd have five instead of six arrows (one would be bidirectional) and we'd have only one arrow-crossing instead of two.

Stefan, would you like me to change this? It shouldn't be too hard

 

[Godmil]

What exactly changed your mind?

Sarah's second set of 4 images. In the 4 cases one could be solved with a J perm and the other 3 needed AUF's we differed in opinion on which of those cases was which, but there was no difference in the moves. So it seemed there could be no meaningful difference in the definitions.

 

[cubernya]

Actually, in regards to the diagrams on the G perms, it would be harder than I thought. I was just thinking about the parity issues, then pulled out my cube...If you have a 2 swap of corners then it is necessary to have either a 2 swap of edges or a 4 cycle of edges.

 

[Stefan]

Stefan, would you like me to change this? It shouldn't be too hard

Nah... not worth it, plus that shouldn't be decided by one person. I'm sure many people would object, maybe even me, after thinking it through. Was just something I realized and wanted to share.

 

[cubernya]

Just wanted to see what it would look like. I had to put in the 2 cycle of corners and a 4 cycle of edges. It looks a little weird, but it works.

Obviously just a rough sketch. Could redo and make it look good if people like how this appears.

 

[qqwref]

Would you say that combining R' R' into R2' is okay?

Very arguable. Personally I'd say no.

PS: The reasons the G-perms are written as two 3-cycles is because nobody wants to have to recognize one of four edge W-perms. It's just too ugly. (I don't think anyone recognizes G-perms based on cycles anyway, though - seems like the most natural way is to just do the block + two corners style.)

 

[Jaycee]

Very arguable. Personally I'd say no.

IMO, if you're going as fast as people like Sergey (and the other Sergey) are, it'll look like an R2' either way.

 

[qqwref]

Yeah, but I guess if you look closely you can tell whether they are taking advantage of the R+R or R'+R' or not (that is, by using a special grip/fingertrick specifically designed for that sequence of PLLs). As long as they're doing what they would normally do to execute one PLL, and then doing the same for the next one, and so on, it's fine.

 

[Dene]

Nah... not worth it, plus that shouldn't be decided by one person. I'm sure many people would object, maybe even me, after thinking it through. Was just something I realized and wanted to share.

Oi you, clear up some PM space

 

[antoineccantin]

Just wanted to see what it would look like. I had to put in the 2 cycle of corners and a 4 cycle of edges. It looks a little weird, but it works.

Obviously just a rough sketch. Could redo and make it look good if people like how this appears.

I've seen that somewhere before...

 

[cubernya]

You may have, but I just made that diagram in paint, so it's not the exact same thing.

 

[Lucas Garron]

The source code to these is embarrassing. Someone should do this over in SVG; it'd be faster and prettier.