L2L4 Documentation ("Finished") and Discussion

[ben1996123]

Yay I finally know what L2L4 is Seems like a really good method, I may end up learning it. But 220 is a lot of algs...

 

[somerandomkidmike]

I think Kirjava did - but how to do CLL recognition on R when my hand covers the R Layer? I also end up in the high 40stm / low 50 stm though I believe the my penalty for not knowing all Watermann EO cases should be only 4 moves and for not knowing watermann optimal CLL mayby 1 move penalty.

Also the fact that watermann is rarly used, makes me believe that a more mainstream compatibel system, could become more used even if it requires faster turning.

Although I don't doubt the potential of L2L4, I don't see how it would be a more mainstream compatible system than Waterman.

Can you explain what you mean by "more mainstream compatible"?

 

[BigSams]

Mmm very symmetrical method, I like it. So nice how as you solve an F2L edge, a bit of LL is done each time. Not too many algs either, compared to some other alg sets. This is history happening before our own eyes. Did any person/people in particular discover and announce the idea first?
Didn't want to comment yesterday before thinking about it properly, but lets see:
I'm personally ambidextrous in cubing so L moves are quite acceptable. But there are many cubers who have a very dominant - let's say H for either right or left - hand in cubing. Those cubers could do the 1x3x3 block on R or L, and use mostly H, U, F,D moves for all four algorithms. Might even get a better view of the relevant pieces that way. I also find x cube rotations easier than y, or z, and most rotations during algs and the 180 turn between algs 2 and 3 woul be x, x' or x2 this way.
@ stachu, Speaking of the 180 turn, I thought you said you were making it a d2? Much better than y2, and if the above variant were used, an h2 wideturn would be used (meaning r2 or l2) - even better.
If you're using a CLS hack, you are being considerate of the other 3 F2L edges and the orientation of the LL edges... since you're using CLS on slot 1 even though it was designed for slot 4, doesnt that waste moves?
anyway, I really like the pure method, it was cool of you to develop it. BTW, what does L2L4 stand for?

 

[StachuK1992]

Yay I finally know what L2L4 is Seems like a really good method, I may end up learning it. But 220 is a lot of algs...

I don't plan to learn that many. I plan to learn 95. Knowing CLS really helps here, I guess.

Although I don't doubt the potential of L2L4, I don't see how it would be a more mainstream compatible system than Waterman.

Can you explain what you mean by "more mainstream compatible"?

I also don't think it would be very mainstream, at least for a few years; it's just too many algs for most people to try learning.

~quoting by section:~

"So nice how as you solve an F2L edge, a bit of LL is done each time."
Yes, it feels rewarding; it looks to others like you're just placing an edge in, but you know it's more than that.

"H, U, F,D"
Hrm, I'll look into the use of these; I was thinking about this in regards to CP mainly.

I will do a d2. I'm hoping people will have the intuition to do whichever is faster for them.

"If you're using a CLS hack..."
I'll optimize the nasty cases.

I've no idea of the history or naming.

 

[oll+phase+sync]

History:

L2L4 = Last 2 Layer in 4 steps it was first developed by Duncan Dicks early this century

Promotion

If you don't like doing y2 during the solve do d2 or u2 since the method does not require you to Sync your D - Layer all the time (but I currently feel better with syncing D and centers)

I call it 'mainstream compatible' because it is easy to describe, easy recognition, alg count can be reduced by skipping / avoiding unlikely cases.
Although if someone would post a sub waterman time with L2L4,
it would boost L2L4 much more, then posting a new watermann record, would boost watermann popularity.

EO can be don intuitivly <- I should add more of these but I'm in a hurry


Intuitiv/iterativ EO

if you hear iterative it's aparent I'm doing something like Roux EO or more precisly the same as Kenneth on his L5E wiki page

Summery of the system:

If you don't know an EO alg
1. change target -> you now whant to place FR edge (this requires no moves, but make sure you know wich edge you want to send to DF later on with the final M )
2. move an oriented edge (not FR itself ) to UF position ( 0-1 move)
3. F (1 move)
4. AUF + M'U2M ,AUF + M'UM, AUF + M'UM, AUF + M'U2M, <- this is no algorithem but a building block notation you never need all 4 blocks like noted here , 3 is the maximum , 2 the average.
5. AUF + F' + y

Please note after doing the starting F turn the definition of orientation for a middle layer edge will flip (Yes, the definition flips not the edge ! It becomes aparent if you try it out.)

Please note also that successive M slice turns must always alter direction to preserve DB edge. (this is the only difference to Roux EO)

I'm sorry I have no more time now to explain further, (Roux User should already know everything)

This is in fact my sedcond try to get the iterator thing running (the frist try had an akward ending, but that's fixed now as you see)

When I'm getting some spare time, I will compare all EO cases : how good the iterator is. And add some tweaks (if I find some)

If someone believes he can discribe this better - please do so

 

[irontwig]

I call it 'mainstream compatible' because it is easy to describe, easy recognition, alg count can be reduced by skipping avoiding unlikely cases.

Doesn't it make more sense to call Waterman more mainstream since it shares more algs and concepts with CFOP/CFCE/Petrus/Roux/CF?

Although if someone where to post a sub waterman time with L2L4 it would boost L2L4 much more then posting a new watermann record would boost watermann popularity.

lol, what "Waterman popularity"? There's like one person using Waterman and he doesn't even know all algs yet.

 

[oll+phase+sync]

Doesn't it make more sense to call Waterman more mainstream .

lol, what "Waterman popularity"? There's like one person using Waterman and he doesn't even know all algs yet.

with Stachu and me that makes two users for L2L4 , means it is twice as maintream as waterman - (I'm really sorry for this polemics , but I can't resist)

For the second You are right, it's worthless to mention a method could be faster than waterman, instead I shoul compare it to some popular method and state that was faster than 90% of that methods users could do (that would be even more polemic )

 

[Zubon]

I would love some more example solves.

Especially ones showing tricks to quickly do the first layer with block building.

 

[StachuK1992]

I would love some more example solves.

Especially ones showing tricks to quickly do the first layer with block building.

Sure, although I warn you that I'm not great.

 

[irontwig]

I would love some more example solves.

Especially ones showing tricks to quickly do the first layer with block building.

I threw together three example, needless to say I'm not too used to this step:

U B2 F' D2 B L' F2 B R' D' F' D2 F' B U' B2 F L' R B' F R' F' D' R2
R E2 F' L' B u' R2 B R B' x2 M' U r' U R U' R' U R (19s)

D' F2 L2 B L' R F2 B' L2 U2 R2 B R L D2 R' D' F B R F B U' B' U
y2 R B' U' R' D2 r2 R' U R y U2 R U R u R' U' R' U R M U M' (22s)

F2 R B D' R' F' U' R2 B R' B' D U' F' B R' L2 U L F2 U2 B' L2 R B
F x' U R' F U2 F R' x' y R2 U' R2 S' U2 S (15s)

A rouxer could probably do better.

Edit: Just two more:

U2 B F2 U R2 F L2 D' F U2 L2 B2 R F2 L2 U2 R D2 L2 R2 F2 B2 U2 D F
U2 r2 L U L' U2 b' R r2 F L' z' M' U2 M (14s)

B2 R' L2 B' F2 L D2 R2 L U' B2 F' R' D R' F B U B' R F U D B2 R2
B' L' U' F R2 F' U2 F2 z2 y U2 R' S' U S (13s)

I forgot to say that these are just the first five random scrambles I got from qq and they're all on U.

 

[StachuK1992]

F2 D F2 B2 R2 D' U F2 R2 U' F' B' L R' F2 D2 F B R' B D R2 U' F' U2
1x2x2: F L F2 (3)
1:2x3: R2 U' M r' U x' (5/8)
1x3x3: U' R2 U' R U' R' U (7/13)

F2 B L B2 U2 R2 B R2 B' R F D' R2 F2 L R2 D' B2 U F2 U R2 L2 D B
y R' U r2 R D2 (5)
x z L M' U M B' (5/10)
l' U2 L' U L' U2 (6/16)

U' B2 F' R2 F' R' U2 R' L' B2 U2 D2 F R' U2 F2 R' F2 L F' D2 U L F' B
D2 R2 L' D (4)
x' z
U M' U' r2 U R2 U' (7/11)
x
r' U2 R2 U2 R2 U R' U R U' (10/21)
Ugh pathetic.

more shall be posted later.

 

[jms_gears1]

Ill do some solves later, Although i suck at block building... Im probably better than you ;P (or not lol idk)

 

[StachuK1992]

Movecount-related stuff

According to my documentation:
CO = 9.1 moves
CP = 12.2

EO, EP 'pure' = 12.2 + 11.9
FL+FR, ELL = 11+11

The first layer should be 15 moves or less by human, I'd suppose.
So essentially,
Pure = 15 + 9.1 + 12.2 + 12.2 + 11.9 = 60.4 moves
Alternate = 15 + 9.1 + 12.2 + 11 + 11 = 58.3 moves

Hrm, thoughts?

Be aware that I do NOT have the best algs.

Also, this isn't like F2L or a block-building method where you have to look ahead tons and work stuff out as you go; these moves will be pre-programmed into you, so will take less time on average.

Personally, I'm liking the alternate, as ELL has uses outside of this method as well.

Edit: where's Kirjava? I had expected him to post here already!

 

[oll+phase+sync]

Movecount-related stuff

Besides feeling that many algs can still be tuned, I feel that your way of "counting" does not include probability (would that higher or lower the movecount), and it doesn't iclude shortcuts (there is no common sense for them at the moment).

Regarding ELL I think knowing all fast ELLs (9 or 10) (like all edge 3-cycles, all oriented) is stongly recommend for L2L4 anyway, (two of them appear even in OLL, 4 in PLL)

though when I reach the L2L4 - EO step (6 edges left) to force an easy ELL I should place FR+FL+ "one more edge" (EDIT someone out there knows how to do this, without raising alg count ?)

So my personal strategy is a little less restrictiv like: just FL + "any two more edges" (as a shortcut case) that will give you an edge 3-cycle at the end (one of the fast ELL's but one edge starts in E -Layer)

 

[StachuK1992]

Besides feeling that many algs can still be tuned, I feel that your way of "counting" does not include probability (would that higher or lower the movecount), and it doesn't iclude shortcuts (there is no common sense for them at the moment).

Regarding ELL I think knowing all fast ELLs (9 or 10) (like all edge 3-cycles, all oriented) is stongly recommend for L2L4 anyway, (two of them appear even in OLL, 4 in PLL)

though when I reach the L2L4 - EO step (6 edges left) to force an easy ELL I should place FR+FL+ "one more edge" (EDIT someone out there knows how to do this, without raising alg count ?)

So my personal strategy is a little less restrictiv like: just FL + "any two more edges" (as a shortcut case) that will give you an edge 3-cycle at the end (one of the fast ELL's but one edge starts in E -Layer)

You're right, it does not include probability. These were meant to just be estimates.

I'm not sure if I like the half-and-half FL+Whatever strategy. L2L4 was meant to be a fairly static method, and I believe that this might harm that.

OLL has sent me some nice CP algs; I'll be publishing those and some more information tonight. Thanks!

 

[jms_gears1]

U' F' U2 R F2 L2 U2 B2 U2 L F2 D2 L D F2 B R2 F' R' L U F2 B' L' R2

x'yD2FUBR'BUrB'UR2U2R'UR'U : 16
z' U2 F L2 F' L2 F' L U L' : 9
y L U L U L U' L' U' L' : 9
y U2 F' R2 U' R2 F2 R' U' F U2 R U2 F2 U R : 15
R U' R2 U2 R2 U2 R U R' U' R' U' R2 U2 R : 15
U'D : 2

I love the feeling of this method.

EDIT: lol 64 moves. i had the worst EO case >.>

 

[jms_gears1]

If I am right, you should only actually need to learn 2 of the cases for EO. Because AUF doesnt matter.

 

[StachuK1992]

If I am right, you should only actually need to learn 2 of the cases for EO. Because AUF doesnt matter.

lolwat.

 

[Kirjava]

I love reading about new methods and I think you've done a great job!

Not new.

However in the time it takes me to solve one layer, I could just use F2L and insert the edges at the same time. For me, there is not much difference in time between solving the first layer and first two layers using CFOP.

This is meaningless information.

So this method is useful for people that would rather do 2 more algs than insert CE pairs instead of corners.

You could also say that it's useful to people who like the method.

... so in contrast to stachu I prefer to skip an edge in first layer.

Pseudosandwich method.

But just dealing with on color should make this kind of blockbuilding much easier than for example Petrus

A layer is more awkward than a 2x2x2.

Personally I also feel that doing L2L4 color neutral is much easier than F2L or waterman color neutral.

Why?

Has anybody thought of doing CLL on R rather than U for Waterman thus saving one or two rotations?

Won't save any time really.

I'm personally ambidextrous in cubing so L moves are quite acceptable.

Right handed people never use algs with L moves in.

Edit: where's Kirjava? I had expected him to post here already!

Hey baby <3 Want me to do some more example solves sometime?

 

[StachuK1992]

Hey baby <3 Want me to do some more example solves sometime?

That would be lovely! Really, I'm looking for FL examples. After that, it's fairly straight-forward (unless you can use a nifty way to MOO L6E? I know I mentioned it earlier in this thread, after OLL (I need something better to call you!) mentioned it in the other thread.)

OLL has already sent me some algs (which are much appreciated!) that I'll will be updating the list with before too long.
Also, a nice wiki page is needed, so I'll work on that in the coming days.