Do any CFOP cubers solve the 3x3 without any cube rotations?

[TDM]

I agree, I don't use the rule to reduce rotations. It's just so I know that I could find something better. For example set up the cube with R' U' R U R' U' R U2 R' U R then think about how you would solve that case normally then try the inverse R' U' RU2’ R' URU' R' UR
Most people I have seen would rotate and end up having to rotate again to solve that mostly because it can be hard to see exactly what case it is but since it is a good edge I was able to find a way to do it without rotating and saving a rotation that time was worth it for me because before it was 2 rotations for 1 case.

I'd only do that alg to preserve EO; otherwise I'd do two rotations. That case is the only case where I would almost always choose to do the rotation. I'd agree with you for any other case though.

 

[Robert-Y]

Avoiding rotations for the sake of avoiding rotations is stupid. Solving a particular case in the fastest possible way, being it a rotation, or not a rotation is not stupid.

Thank you!

I don't see anyone asking similar questions like "do any CFOP cubers solve the 3x3x3 without any L turns?"

What's the difference? A rotation and L are both considered turns. Both can be avoided completely. Why not consider my question above as well? Is it a stupid question?

 

[Schmidt]

What's the difference? A rotation and L are both considered turns. Both can be avoided completely.

not in the same solve. If you have completed the R side of the cube, how would you solve the rest without L, y, z or d moves in a speed solve?

 

[~Adam~]

iirc Cornelius Dieckmann had a video of a very fast average rotationlessly a couple of years ago.

 

[Robert-Y]

not in the same solve. If you have completed the R side of the cube, how would you solve the rest without L, y, z or d moves in a speed solve?

Well my first question was only about L turns, so the solver is free to do rotations.

 

[mark49152]

Yes it does. It tells you to rotate when it's faster than not rotating.

"Informative" was the operative word there. Your statement is true but not informative because it's already obvious to anyone who grasps that the point of speedcubing is to solve it fast.

Here's a few random thoughts and opinions on rotations and when they might be good or bad, some of which are probably also obvious to many.

1. Rotations aren't bad in themselves. They have a bad reputation because they often cause re-grips, and re-grips tend to be slow and therefore bad.
3. Re-grips sometimes arise for reasons other than rotations, so minimizing re-grips might be a better way to think of the challenge. For example, switching between solving on L and R can result in re-grips.
5. Technically, d and d' turns are rotations, but sometimes they can be easier to execute than full rotations as they might not require as complete a re-grip.
7. Other things can often be slow and bad too, like pauses or B moves or long or awkward combinations of moves, and sometimes even a slow and awkward rotation might be faster than the alternatives (which I think is what applemobile is saying above).
9. Rotations aren't a black and white case of either you're turning the cube by 90 degrees or you're not. Some rotations can be natural, flowing movements where the re-grip and the next turn start before the cube has even rotated half way. Other rotations are more awkward.
11. Like any other cube movement, rotations can be faster when practised and committed to muscle memory as part of a sequence of moves. That happens when rotations are included within algs. For F2L, rotating just before inserting a pair is a sequence that's likely to get practised a lot, since most F2L cases end that way. Rotating to align an F2L case before starting execution may be a less familiar sequence, and so that kind of rotation might be slower due to less practice.

 

[Escher]

Rotations are just another word for a movement of the cube. If the movement of the cube is unnecessary either for immediate benefit or for future investment in speed/efficiency then it's a bad idea.

'Good' solves can have 3-4 rotations in, equally, terrible solves can have 3-4 rotations in. Don't be mislead by nomenclature!

 

[GuRoux]

about how many rotations do top cfop users use.

 

[ryanj92]

about how many rotations do top cfop users use.

Went to the 3x3x3 WR average... Most obvious source I could think of

not including PLL rotations.
solve 1: 1 rotation, 2 d turns (basically hidden rotations)
solve 2: 1 rotation, 2 d turns
solve 3: 1 rotation
solve 4: 2 rotations
solve 5: 1 rotation

 

[GuRoux]

Went to the 3x3x3 WR average... Most obvious source I could think of

not including PLL rotations.
solve 1: 1 rotation, 2 d turns (basically hidden rotations)
solve 2: 1 rotation, 2 d turns
solve 3: 1 rotation
solve 4: 2 rotations
solve 5: 1 rotation

thanks. so i guess it's about 2, counting d as rotations. Though d turns aren't too bad if there needs no U moves after.

 

[caters]

what I often do is first the algorithm that has rotations without rotations(I mean if the Z and X axes are both horizontal how do you know which is X and which is Z plus when I do a y rotation so that yellow becomes my F side I never get it in the right spot), do as many algorithms that have rotations without rotations as I need until I come across an algorithm without rotations and than do that rotationless algorithm.


Another problem I have with some of the algorithms is that it either messes up my F2L(that is after I am done with the algorithm a pair at least is out of place and usually a whole section of my cross) if I am in OLL or PLL(more often PLL) and sometimes even when I solve another F2L pair or the solving portion of it as written does not get it in the right spot and makes me have an edge down, corner on top case if I am in F2L.

 

[Schmidt]

It seems you got the notation wrong.

if you do a F move but with the whole cube you have rotated z

F = z
F' = z'
U = y
U' = y'
R = x
R' = x'

and while we are at it, middle slices.

M = middle layer goes in the same direction as L
M' = L'
S = B
S' = B'
E = D
E' = D'

http://www.speedsolving.com/wiki/index.php/3x3x3_notation

 

[mark49152]

S = B
S' = B'

I believe S follows F, not B. The inconsistencies in our notation are so silly and annoying.

 

[Tianyu Xiong]

ZZ method?

 

[mark49152]

ZZ method?

The discussion is about CFOP.

 

[brian724080]

^^

ZZ method?

Plus, sometimes rotationless EOLine is no good, and the same with LL.

 

[Petro Leum]

^^



Plus, sometimes rotationless EOLine is no good, and the same with LL.

actually, for two handed i (almost) never get rotations. my f and b moves are pretty good, and i dont need rotations for COLL. plus when you know your color schemes, you can predict most cubies so you dont have to look around on other sides than U/F/R/L

unless i get OLL skip and have to do PLL, then i might need some in the algs