Should We All Relearn PLL?

[Lofty]

People have thought and have run all OLL and PLL through a cube solver. And if for F2L you just mean the 41 cases then yes they have done that too.
Optimal algs are often quite nasty to execute and contain random moves all over the cube that are not fun to execute at all.
I myself would love to learn RFU cases for all my PLL's.
And a lot of people here don't just search around on pages they make the pages.

 

[Stefan]

M2 U M E2 M E2 U' M2 ... 8 Stm

Dude! Highlight my message you replied to!

Ok, I can do that in the future. I normally do not when the post I'm replying is immediately above the one I'm writing.

Not quote. Highlight! So that you can see my hidden message!

 

[Stefan]

I think that is a little bit off Stefan.

Well, I feel how I feel.

If you read my first post I said that I took a cubing break so I had no idea what was going on in the speedcubing world really.

Yeah but you said the break started last year. And we've been optimizing algs with computer help, including shortest algs, at least since 2003.

The algorithms reduced many of the usual moves required to solve each PLL case, turning a 13 move algorithm into a 9 move algorithms as well as reducing all the others by several moves.

Only if your personal selection of algs truly sucked. Shortest PLL algs have been known, and for many cases used, for years!

I mean, if you can turn 2 tps, [...]

For a PLL? Are you talking about total noobs?

 

[Stefan]

Oh and only the edge-3-cycles and the corner-3-cycles have 9 move algs. If you used 13 moves for either of those, your alg selection was even worse than I thought. I never heard of anyone using 13 moves for those cases, and the 9 move solutions have probably been known since the early 1980s.

 

[Kenneth]

Not quote. Highlight! So that you can see my hidden message!

Ehm sorry, don't know how to do that, something I must have missed :-(

 

[Stefan]

Simply drag your mouse cursor across the message text. But ok, I repeat it here:
" In HTM, of course, don't be a nitpicker like me."

 

[Kenneth]

Aha, that type of highlight =)

Was afraid you should say something about HTM after I posted my alg

 

[AvGalen]

Isn't all of this a complete misrepresentation of the order algs were found?

I think this is the correct order:
1) Algs were found by playing around and using commutators and setup-moves. These algs were mostly long and slow, but some nice ones were already found this way
2) Algs were improved for fewer moves by hand and some use of computers
3) Optimal algs were found for every case (mostly by use of computers)
4) Sub-optimal algs were found that were faster to execute (mostly by handpicking from a list of optimal/near-optimal algs, but also by combining common fingertricks or picking from a generated list of RFU-only moves)

Or short version:
1) Find an alg that does the job
2) Find an alg that does the job better
3) Find an alg that does the job in fewest moves
4) Find and alg that does the job in fastest moves (might require more moves)

The best example I can think of is a H-perm
1) (M D2 M') U2 (M D2 M') U2 (M D2 M') U (M D2 M') U2 (M D2 M') U2 (M D2 M') U'
2) (M2 U2 M2) U (M2 U2 M2) U'
3) U (R2 F2 B2 L2) D' (R2 F2 B2 L2) (10f*)
4) (M2 U) (M2 U2) (M2 U) M2

Another example would be U-perm
1) (M D2 M') U (M D2 M') U2 (M D2 M') U (M D2 M')
2) F2 U' (M' U2 M U2) U F2
3) F2 U' L R' F2 L' R U' F2 (9f*)(basically the same as 2))
4) R U' R U R U R U' R' U' R2

And even right now, faster algs are still being found. Also, algs are being optimised for one-handed moves and even for big-cubes (R2 U2 R U2 R2 U2 R2 U2 R U2 R2)
And even algs that are already RU-only can be improved. R2 U R2 U R2 U2 R2 (U2) is slower for me than (R U R' U')(R U R' U')(R U R' (U'))

 

[Jai]

Who uses these algos? I do. I changed my F perm, N perm, and R perm to one of the RFU optimal algos.

Would you mind sharing which ones you're using? I've never really taken the time to look through all of those algs before .

F perm: U R2 F R F' R' U' F' U F R2 U R' U' R
R perm: (it's the common one) R' U2 R U2 R' F R U R' U' R' F' R2 U'
N perm: R U' R2 F2 U' R F2 R' U F2 R2 U R' U

 

[Harris Chan]

How about R2 U (F B') R2 (F' B) U R2 = R2 U S' U2 S U R2 = 7 STM?

Edit:
You can see Jason Baum using this alg here (it might be the inverse but you get the idea):
http://www.youtube.com/watch?v=OEt1yRRQ7Mw

It's Matt Walters signature move ;-)

 

[Speedy McFastfast]

M2 U M E2 M E2 U' M2 ... 8 Stm

I love you.

I'm not making any promises, but I think that alg might actually work better for me than the one I'm using now Even other people's arguments teach me nice new tricks!

 

[Crzyazn]

Like i really want to learn an Nperm that's even HARDER to get fast......

 

[Kenneth]

Arnud: I think Ra U2 Ra' Fa' U2 Fa goes for H-PLL number 3 and 4. It is ten turns HTM but is executed as six turns.

 

[Dene]

4) R U' R U R U R U' R' U' R2

Ok so I was playing around with this a bit. It seems to be a mirror or inverse or whatever (can't be bothered to think it through) of R' U R' U' R' U' R' U R U R2. I may just be showing my ignorance, but certainly the one I posted is much more finger-friendly? I'm now waiting for a top cuber to come and tell me that they execute your one sub-1, so damn you . The R U just isn't nice to perform and can only be done with the right hand, whereas R' U' can be done with right wrist, left finger, which seems more comfortable to me.

 

[Lofty]

I use both of those algs Dene Depends on how the cube hands me the U and I can avoid cube rotations and AUF.

 

[Stefan]

I think this is the correct order:
1) Algs were found by playing around and using commutators and setup-moves.

The first way I found algs was *not* with commutators but with the "destroy what you have, repair it a different way, observe the side effects in the unsolved part of the cube"-principle.

 

[FU]

I think this is the correct order:
1) Algs were found by playing around and using commutators and setup-moves.

The first way I found algs was *not* with commutators but with the "destroy what you have, repair it a different way, observe the side effects in the unsolved part of the cube"-principle.

He did mention 'playing around'

 

[AvGalen]

Arnud: I think Ra U2 Ra' Fa' U2 Fa goes for H-PLL number 3 and 4. It is ten turns HTM but is executed as six turns.

Can you translate that to RUFLDBrufldbMES notation?

 

[Stefan]

http://www.ws.binghamton.edu/fridrich/Mike/notation.html

 

[AvGalen]

Ra U2 Ra' Fa' U2 Fa = RL U2 R'L' F'B' U2 FB

I know that alg for One-Handed. It is probably discovered at "stage 2"