2
$\begingroup$

Very often I get situation like this:

Four flipped edges in the equator layer

I am interested in manual solving, so automatically generated solution is not useful (like with Kociemba algorithm). Currently I use twice the two edges flip (E R' E R' E R2 E' R' E' R' E' R2), but this is too long solution with total of 28 moves.

Is there a shorter algorithm to solve this situation?

| improve this question | |
$\endgroup$
  • $\begingroup$ Hmm, what solving method do you use that you get into this situation? (When you solve layer by layer you would never get this case.) Do you use a corners-first method, or a middle-layer last method? $\endgroup$ – Kevin Cruijssen Oct 28 '16 at 22:06
  • $\begingroup$ I do not know, probably corners first seems likely. I have figured out 2x2x2 first, and now apply it on 3x3x3. $\endgroup$ – Evil Oct 28 '16 at 22:11
  • $\begingroup$ @Evil. You counted both quarter turns and slice turns as 1 move. But in slice turn metric (stm), any layer is only counted once whether a quarter turn or a half turn, and R, R2, E, and E2 are all one turn. - So, with the slice turn metric, your formula is really length 24. Jonathan's formula is of length 12, and Jaap's formula is of length 14 but easier. - Yes, this occurs with the corners first method. Have you checked out the Varasano method? It solves 8 corners in 3 steps: 1. orient first face (intuitive), 2. orient opposite face (7 cases), 3. pbl (position both layers, 5 cases). Done. $\endgroup$ – Cuc Apr 3 '17 at 5:12
  • $\begingroup$ Wow. Thank you @Cuc. I had apparently counted in QTM. I never heard about Varasano method, but honestly I have never learned any known solving system, as I tried to generalize from 2^3 to 5^3 on my own, then picked some algorithms to speed up solving, trained fingers for speed. Now it is probably good moment to learn something, so I will definitely learn Varsano method. It seems far faster with only 12 cases, because right now I use more with slow 2-3 swaps. Thank you again for pointing this out. $\endgroup$ – Evil Apr 3 '17 at 18:31
3
$\begingroup$

The shortest number of moves in face turns (HTM) from here to solved would be

$16$, for example: U R F' D L2 B' R U' D B' L F2 D' R B' D'

The shortest number of moves in slice turns (STM) from here to solved would be

$12$ (matching yours), for example R U2 R2 S R' E2 R' U2 R2 S R E2

I don't do speed solving so maybe others can give some advice, but I think using some look-ahead or a different approach that avoids this may be worth looking into. Also note that a longer sequence might actually be faster to perform.

A method that is probably a good start might be the one made by Lars Petrus (also a good start for fewest moves), which may be found at his website.

| improve this answer | |
$\endgroup$
2
$\begingroup$

One simple 4-flip sequence that I like a lot is (E R)4. It doesn't quite flip the four edges you want (it does FL BL BR and UR). A small conjugation to to bring the FR edge to UR then gives the following sequence:

F' U' F (E R)4 F' U F.

It is not quite as short as Jonathan's, but it is easy to remember and understand.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Wow, thank you this one is also very cool and intuitive. $\endgroup$ – Evil Oct 30 '16 at 19:24
  • $\begingroup$ Oh, so "the basic case" is for this, with additional moves to put top edge to the equator. $\endgroup$ – Evil Oct 30 '16 at 19:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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