[Help Thread] CPLS and 2GLL discussion

[Baian Liu]

I didnt understand what EOCPLL does.. If you are just permuting the corners should be only 2 algorithms, and with less move count.

EOCPLL is edge orientation and corner permutation of the last layer.

 

[StachuK1992]

EOCPLL is edge orientation and corner permutation of the last layer.

See the CPEOLL page on the wiki.
Except expect better algs for what Baian is talking about - he doesn't care about corner orientation, while the algs on that page *do*.

 

[Cyrus C.]

Stachu generated EOCPLL algorithms when making Seth since it's a subset. Seth algs are here, look on the top rows of UB UL, UF UB and UF UL UB UR.

 

[DaKrazedKyubizt]

oh, I am SO learning this!!!

 

[Baian Liu]

Some examples of EOCPLL:

R U2 L D R2 B L2 F2 U2 R L' F U' R' B D' R2 F R L' U2 D F U2 D2
Cross: z2 L' F' D F D2 F (6)
F2L: U' R' U' R2 U R' + U F R' U2 R F' + U L U L2 U' L + Dw R' U2 R U' Lw U' Lw' B (OLL skip...) (33)
EOCPLL: F (R U R' U')x3 F' (47)
2GLL: U' R U R' U R U' R' U R U2 R' U' (60)

R D L' D2 B F D2 U2 L' B2 L2 R2 D' U' B L B D2 U' F2 B R F2 B2 L2
Cross: z2 F D F D L' (5)
F2L: R U' R' y R' U' R + R U' R' y' L' U' L + E' R U R' E + R U R' U R U' R' (27/29)
EOCPLL: F R' F' R U R U' R' (35/37)
2GLL: R' U' R U' R' U2 R2 U R' U R U2 R' U' (49/51)

U' F2 U' D' L' U2 B' L F' R' U' D' R B L' R2 D2 F D L' B2 L' D B F2
Cross: y' U' R' L U' Rw2 (5)
F2L: U' L U' L' U2 L U' L' + R U2 R' Dw L U L' + R' U' R + y L' U L U F U' F' (30)
EOCPLL: L' U R U' L U R' (37)
2GLL: U R' U' R U' R U2 R2 U' R2 U' R2 U R U' (52)

Stachu generated EOCPLL algorithms when making Seth since it's a subset. Seth algs are here, look on the top rows of UB UL, UF UB and UF UL UB UR.

Ah, I didn't know about Seth. Thank you for the algorithms. I have hand-found my own already though.

 

[Diniz]

Really cool, the movecount is lower then OLL+PLL right? I might try to find some OH algs later..

Edit: A better EOCPLL in the first exemple would be R' U' R U' L U' R' U L' U2' R (11)

 

[Baian Liu]

EOCPLL algorithms

 

[StachuK1992]

Apart from the "Done, all flipped" case, really nice!
I'll post some algs I've been using here in a bit, after some comparisons.

-statue

 

[StachuK1992]

Some alg suggestions for EOCPLL:

I'm rather biased here, but the <MU> pure 4-flip is faster for me than your crazy z U R x' U' D2 z' x' U' R' U Rw U' R' y' R

Actually, I don't like any of your 4-flips. They're too rotationy for me. I'll refresh this post when I decide on better ones for me.

 

[Diniz]

Some alg suggestions for EOCPLL:

I'm rather biased here, but the <MU> pure 4-flip is faster for me than your crazy z U R x' U' D2 z' x' U' R' U Rw U' R' y' R

Actually, I don't like any of your 4-flips. They're too rotationy for me. I'll refresh this post when I decide on better ones for me.

I think they are that way because they are OH algs..

 

[Baian Liu]

I think they are that way because they are OH algs..

I made it messy to discourage people from using them. Because basic partial edge control can prevent these cases.

 

[Diniz]

I made it messy to discourage people from using them. Because basic partial edge control can prevent these cases.

I srsly need a way to avoid non oriented case in OH, the way i used to do is to slow =/

 

[hic0057]

Bump and slightly off topic:

I've been trying to make a method for 2x2 base on ZZ-d (If it turns out to be a success it is named Hickingbotham Method)
Steps:
Create a corner permuted Corner-corner pair turning it from 3 gen to 2 gen
Solve the rest 2 gen

This is how far I got of learning it

Create a corner pair, This is usually already made
Solve the 2 corners that needs to go next to it: Orientation doesn't matter
Recognise the Corner Permutation:

If it is an adjacent swap have the ones that need to be swapped on the Right and perform F R' F'
If it is a diagonal swap then perform F R2 F'

Any suggestion, ideas, tips, etc are most wanted

Thanks

 

[StachuK1992]

Here you go.

 

[oll+phase+sync]

I've been trying to make a method for 2x2 base on ZZ-d

For my taste "your way" and also "brass"-way has to much setup moves. I would prefer a way like original zz-d where 2-gen reduction is done, while building the first pair.

That way I would have full 15 sec to do the reduction analysis, sadly I have no concept how to do this analysis.

 

[StachuK1992]

For my taste "your way" and also "brass"-way has to much setup moves. I would prefer a way like original zz-d where 2-gen reduction is done, while building the first pair.

That way I would have full 15 sec to do the reduction analysis, sadly I have no concept how to do this analysis.

I have some ideas.
Give me a day.

 

[mariano.aquino]

i just discovered this thread. i´ve also been doing research on this CPLS concept, but, with a different perpective. yet, there´s nothing concrete. could it be possible to determine if corners are ok or not directly from inspection...?
my approach is realy similar to the 2x2x2 version shown here, almost like a corners first, but u would only care to build two corners (DLF and DBL) and permute 2 of the other 6 corners left, to leave them in 2gen. then continue with either roux, or a wierd version of fridrich i came up with, only using R, M, r, and U. for OLL , there´s an already created set, that i found under the name of SuneOLL, (yau-kirjava) which is OLLs only using variants of sunes, thus, using only those same generators, and of course, then just EPLL.
even though i cannot ensure this is "speed-friendly" it´s really fun, and u can get at least sub20, without much effort.
however, i do think, if the pattern can be actually recognized from inspection, this could totally benefit Roux users, providing a new variation, and this is a big item to point out
i´ll keep working on it. like someone said up here in the thread, if this method works, it shall be called ...F2G (Full 2-Gen Method) or FRUM variation (Full R-U-M) or simply MA (my initials).

I leave you with a sample solve, so as to depict the whole method best:

scramble:
L' F' B U' D' B2 L B' D' L' F' B D L2 B D2 U2 F' D' B D2 F' D' U2 B2 (25htm)

xz' L'U'L y R2 U + F R'F'(CP recogn + swap) 8/8m
M E'M u (1x2x3) 4/12m
r'U'r (2x2x3) 3/15m
R'U R U R'U'r U R'U'r' (3rd slot) 11/26m
U'R U M'U R'U'M (F2L) 8/34m
(U) R'U'R U'R'U2 R (U') r U2 R'U'R U'r' (SuneOLL) 14/48m
M2 U M2 U' E2 M'E2 M' (Z PLL) 8/56m

(+2 AUF)

it´s just my first approach
Main uses?
OH, maybe 2H (for fun, yet)
with Roux, this would be even faster, since u start with ur 1x2x3 block, build 2nd block with no limitations, and ur CMLL is reduced to just 7 cases, that u could easily combine with 7 COLL cases, and increase the number of skips.

can u figure out which permutation u need to do in 15 seconds..?

furthermore, would u get used to "no-inspection solves" in exchange for an absolute 2gen solution?

i´ll be updating when something new comes up =)

comment! anything could help!

 

[Petro Leum]

Seems extremely nice. But (maybe im blind) i find now real explanation on how it works (when to use the algs, how to recognise them).

 

[samchoochiu]

Sorry to bump an old thread, but I really wish to find a full page of 2GLLs. Most of the 2GLL links on this thread and the wiki don't work and others give me a 404. If someone could help me out I would really appreciate it. Thanks!

 

[antoineccantin]

You can find them all on boca (http://boca.bee.pl/login.php?l=en), however, sometimes even the best alg isn't all that great.

Stachu also has a list of 2GLLs (http://stachu.cubing.net/cpls2gll/2gll.txt).

If you have a case with no good alg, you can contact me, I might know something better for that case.