attention, zz and petrus users, take a look at this method.

[cubeRemi]

hello everyone, here I will post my method to solve the last 2 faces in the petrus or zz method ( the edges are already oriented ).

after 2x2x3 (and petrus users after orienting edges):

- solve a 2x2x1 blok on the right side (petrus 4a http://lar5.com/cube/fas4.html )

- solve the last down corner on the right side ( ignore corresponding edge )
- coll ( orient and permute corners )
- permute edges



the advantige of this method is that you can do a lot of LastLayer corner manipulation, while inserting the last down layer corner.

probably fewer moves

also, while solving the 2x2x1 block you normaly try to trace the next corner + edge. now you only need to trace the corner ( and maybe look into the LL corners already )

the disadvantige is that you wil mis the 1/12 coll + skip.

also the edge permutations may be less well to perform for OH cubing but the algs I present are just more or les easy to find algs, I hope some of you wil find some finger friendly, or fast to execute algs!

the cases I present are all the cases you need to know (I hope)! though you may have to miror or inverse the algs.. so 12 algs, + mirrors and inverse 48 algs
for some strainge reason I could not post the algs in a better way, so open the attach.

please give me comments, sugestions and questions.

Remi

[Swordsman Kirby]

The main disadvantage is that those algorithms for the last 5 edges are horrible.

[ErikJ]

it seems like a good idea, but those algs are not very nice and I don't think recognition would be very good either.

[cubeRemi]

hello,

your are both right, better algs need to be found to speed up. I was just curious what others would think of it.

I don't think recognization is a big problem, if you see 3 edges it is enough to know which permutation case you have... I think coll is much slower to recognize and after getting used to it pll is probably just as hard.

again, if you see 3 edges you can recognize the case...

Remi

[ErikJ]

are the algs you have listed optimal? if so this could be used in FMC I'm not sure of the average number of moves for COLL but it is pretty good. I might consider learning these but I'm unsure of what the pictures show. which square is the cecnter of the LL?

[cubeRemi]

I did came up with this idea for FMC, witch is my favorite discipline!!

if you use optimal coll:
9,78 moves, http://www.ai.univ-paris8.fr/%7Ebh/cube/ =optimal algs.

but in FMC you may be able to do 7,99 since you can do a lot LL corner manipulation..

however, I did some FM comparisons with my normal FM method (intuitive, then 3 corner commutator) , but my intuitive almost always won..
however, it is nice to know this method in case you don't find a nice solution.

I always gave the optimal alg + the next alg..

the first 3x3 squere's are the up layer, 2x3 squeres under that are the front face, perform 2 algs and you'll understand the pics.

from your youtube account ( nice tutorials !! ) I saw your a petrus user,

for FMC count like this
2x2x3 + no bad edges = 13 moves?
2x2x1 6-7 moves?
corner+corner manipulation: 4 moves
coll + ep 20 moves
+1
+= 45 moves

Remi

[ErikJ]

hmmm, I still don't quite understand the pictures. I'll keep trying. but if you are into this kind of thing I recommend that you look into CLS, specifically the I cases.

[Stefan]

B2 turns for permuting the edges? Shouldn't that be <U,R> only? You can use ACube for that.