A Collection of Algorithms

[Christopher Mowla]

Interesting, but definitely not the first. I asked for one a few years ago, and I Mitchell Stern found it.

Which type? <U, R, r>, <U, r>, or <U, Rw, r>?

Fortunately, that's an easy question to answer.

Does this mean you have one in <U, Rw>? If so, how long is it? Or were you implying that I will eventually find one myself?

 

[Lucas Garron]

<U, r>.

And I don't think I have one in <U, Rw>, but I meant it's easy to check if one exists.

 

[Lucas Garron]

@Lucas,
If you for some reason can't find one with ksolve (or whatever your tool of choice is)

Don't have time to read the rest of your post right now, but:

http://www.svekub.se/files/ksolve.zip

 

[Robert-Y]

Megaminx:

R' U' R' U' R F R' F' U R F' U' F U2 R
R' U2' R' U' R F R' F' U R F' U' F U2' R

They're both pure EPLLs which affect 4 edges. (Maybe useful in BLD as well as speedsolving I dunno...)

The first one was found by trying out different <RUF> algs which CE5 generated.
The second one was found by changing the first one slightly...

 

[Tim Major]

Haven't seen this COLL alg before. Also nice, as it's pure (only effects corners)
R U2 R' x (R' F R F')x3 x' R U2 R'
Also quite nice if you execute from the back (starts of as R' U2 l)

 

[riffz]

Haven't seen this COLL alg before. Also nice, as it's pure (only effects corners)
R U2 R' x (R' F R F')x3 x' R U2 R'
Also quite nice if you execute from the back (starts of as R' U2 l)

Why did you add the x moves in?

 

[oll+phase+sync]

Haven't seen this COLL alg before. Also nice, as it's pure (only effects corners)
R U2 R' x (R' F R F')x3 x' R U2 R'
Also quite nice if you execute from the back (starts of as R' U2 l)

I was always to lazy to learn a COLL for this case (EDIT: U+Evelyn+Clix) so your one is quite easy to remeber.
R U2 R' R' F R F' R' F R F' R' F R F' R U2 R' (x-free notation )

The "single" is a nice CLL, too (it's not my AUF angle though )
R U2 R' R' F R F' R U2 R'

 

[riffz]

The "single" is a nice CLL, too (it's not my AUF angle though )
R U2 R' R' F R F' R U2 R'

I actually prefer to execute it mirrored:

R' U2 R2 B' R' B R' U2 R

 

[Kirjava]

rUR'URUL'UR'U'LUM

Just random OLLCPs

 

[Christopher Mowla]

A pure 3-dedge flipper (also rotates top center 180, TMOY!). This algorithm is "2-gen" and it does not use ANY single inner-layer slices!
Rw U2 Rw U2 Rw' U R U Rw U2 Rw' U' R' U' Rw R U R U Rw U2 Rw' U' R' U' Rw' U2 Rw' U2 Rw' ( 36 qtm, 30f)

 

[Christopher Mowla]

Here is a 5 dedge flipper. It is not as efficient as the 3 dedge flipper, but it has the same properties as the 3 dedge flipper.
Rw U2 Rw U' R U Rw U2 Rw' U' R' U' Rw U R U Rw U2 Rw' U' R' U' Rw' U R U Rw U2 Rw' U' R' U' Rw U R U Rw U2 Rw' U' R' U' R' U2 Rw' U2 Rw' (54 qtm, 47f)

EDIT: Here is a single dedge flip, but it too is inefficient. Nevertheless, it has the same properties as the other two.
http://alg.garron.us/?alg=Rw_U2_Rw_...U2_Rw-&cube=4x4x4&notation=WCA&animtype=solveRw U2 Rw U' R U Rw U2 Rw' U' R' U' Rw R U R U Rw U2 Rw' U' R' U' Rw' R' U R U Rw U2 Rw' U' R' U' Rw R U R U Rw U2 Rw' U' R' U' R2 U2 Rw' U2 Rw' (58 qtm, 50f)

 

[TMOY]

A pure 3-dedge flipper (also rotates top center 180, TMOY!). This algorithm is "2-gen" and it does not use ANY single inner-layer slices!
Rw U2 Rw U2 Rw' U R U Rw U2 Rw' U' R' U' Rw R U R U Rw U2 Rw' U' R' U' Rw' U2 Rw' U2 Rw' ( 36 qtm, 30f)

I count three gens, not two (U, R, Rw). Unless you're using the expression "2-gen" in a similarly confusing way as the word "pure" ?

 

[irontwig]

TMOY: lrn2quotationmarks

 

[kvaele]

Alg thoughts

I just kind of stumbled upon an algorithm to cycle the back 3 edges clockwise, and I'm just wondering what you think of it, or have any better algs for it. It is M2 U' M' U2' M U' M2. Execution time is about 1.5 seconds on average.

Sorry, just ignore this thread entirely. Just found it on pll alg list...

 

[vcuber13]

most people already know of this

 

[kvaele]

Wow, sorry... just saw it on pll alg list.

 

[Christopher Mowla]

I count three gens, not two (U, R, Rw). Unless you're using the expression "2-gen" in a similarly confusing way as the word "pure" ?

That is correct. If you have noticed, I did actually find real 2-gen algorithms for the single edge flip. I put the expression in quotes to indicate that it mimics 2-gen. And actually, I think this algorithm type (U, R, Rw) is faster than <U, r> because of the full turn move count.

For example, the algorithm I constructed in <U, r> for the single edge flip was:
r U2 r' U2 r2 U2 r U2 r U r' U r' U2 r2 U2 r' U2 r' U r U' r U2 r2 U2 r U2 r' U2 r' (45q, 31 btm) but it is (64 qtm, 47f). When people say, "Those inner-layer slices kill the speed", that definitely shows how much this algorithm does kill it.

Also, in my opinion, algorithms that have both R and Rw's are faster than algorithms with only Rw's because outer layers are much easier to move, especially on the 6x6x6, which loves to lock up. So <U, Rw> algorithms may not be much better for speed to flip 3 dedges.

 

[whauk]

yesterday i found a great algorithm somewhere in this thread but i cant remember nor find it anymore.

it was either for
U L U L2 F2 L2 U' L' U' B2 R' D2 R B2 or
U R' D' F2 U' F2 U R U2 R' D R2 U R'
and consisted of setup (sexy move)*3 setup'

can s/o help me finding this again?

 

[amostay2004]

yesterday i found a great algorithm somewhere in this thread but i cant remember nor find it anymore.

it was either for
U L U L2 F2 L2 U' L' U' B2 R' D2 R B2 or
U R' D' F2 U' F2 U R U2 R' D R2 U R'
and consisted of setup (sexy move)*3 setup'

can s/o help me finding this again?

I think you're looking for R U2 R' (R' F R F')x3 R U2 R'

edit: Zane_C found this btw

 

[Escher]

I think you're looking for R U2 R' (R' F R F')x3 R U2 R'

edit: Zane_C found this btw

Pretty sure Phil Espinoza posted an identity quite a while ago He showed me it at US Nats at any rate...