The WaterRoux 3x3 method thread

[TDM]

Very similar to a method I proposed a while back, except I did CPFB which let me combine steps 3 and 4 relatively easily. I wouldn't be surprised if your method was better though.
I also couldn't find a good L7E method, so it's good to see you've got something!

 

[Neuro]

I did a small test in regards to movecount and unfortunately it may not be as low as was predicted. It's still about the same as Roux though. Here is my rundown:

FB: 7 moves
Sq+C:11 moves (?)
TCMLL: 10 moves
Setup: 2 moves (?)
L7E: 15 moves (?)

So I'd estimate the average movecount to be 45 which is still excellent and in the range of Roux and ZZ-A, but unfortunately not as powerful as was expected. Correct me if I'm mistaken but isn't this about the same movecount projected in LMCF? I'm thinking that maybe 45 moves is about the est. average movecount that a human method can get right now but that may change in the future.

 

[GuRoux]

i think lmcf is estimated as low 40? kind of interesting because the whole method is 100% algorithms, so theoretically, really fast turning does not increase movecount; giving it an advantage over more blockbuilding methods where movecount suffers due to speed (though blockbuilding becomes pretty algorithmic as well). the question is how good are the moves and recognition. just from a couple rotations, lmcf would probably be worse than roux.

 

[Neuro]

Recognition in LMCF from what I can see could be really fast but would take a while to get used to. From what I see, a lot of the cases (in E2L especially) have poor ergonomics (4/5-gen at points.) I'd say that it's probably on-par with Roux because even though the solves can get very low movecount and a majority of the method is quite fast, the ergonomics of the second step alone can destroy a solve.

What I hope to come out of WaterRoux is a method that gets low movecount solves while preserving high ergonomics and some degree of intuition. But in my eyes, it has a bit of a road block: the movecount is basically the same as Roux so it has very little incentive to switch to unless someone is already a very serious Roux user who wants help with bad cases. Even though it's a good method at the moment, it's either not optimized enough or it needs to be modified to make it feel like a worthwhile method to learn IMO.

 

[crafto22]

I think the big problem with WaterRoux is L7E. It just seems forced. The setup required, the enormous amount of algs and the mediocre recog makes regular L6E a lot more appealing, and the move count is essentially the same. More algs does not mean faster times, especially in a case like this where the setup required for L7E is so great that you might as well just setup L6E after TCLL and you'll get better recog and about the same moves. And without L7E WaterRoux just isn't WaterRoux. So in my eyes this method just won't work out :/

Sent from my Moto G (4) using Tapatalk

 

[Neuro]

Well WaterRoux with L6E is still WaterRoux just with a possibly better algset. WaterRoux with L6E was actually my initial proposal and the way that it's structured gives users an opportunity to choose how they want to finish the solve (L7E, L6E, L5E, or strict Roux)

L7E is efattah's creation and I will support it but personally I will push for L6E to be used as the "main" branch of the method.

What sets WaterRoux apart is in the steps and techniques used, not an algset. That's like saying that using Roux but using Pinkie Pie vs strict LSE is no longer Roux. It's just not the case. It's a different way to finish the solve. Same thing with ZZ variants: there are a countless number of them, but there all just that- variants. It's still the ZZ method!

I don't think that this is a doomed method. Method progression is naturally slow. CFOP has been around since the 80's and we're still finding ways to make it better. It'll probably just be a while before this gets running. I'd be happy to hear any opinions though. I'm still fairly new to method design so it helps a lot to get feedback and I'm sure that efattah appreciates it as well.

 

[crafto22]

Of course, sorry I didn't phrase that properly, what I was trying to get at is that L7E will never work out in my eyes, not WaterRoux as a whole. This is just my opinion of course, but the way I see it, it simply wouldn't be beneficial under any circumstances. Either way, I apologize for my previous statement, I did not express myself correctly. I do strongly believe methods like these are the right path to finding the "holy grail" in terms of methods (and by methods like these, I mostly mean Roux based methods as they seem to have the most potential).

Sent from my Moto G (4) using Tapatalk

 

[Neuro]

All is good, and I understand the viewpoint on L7E. I can discuss with efattah as to what the main variant should be if there even needs to be one for the edges.

I'd be willing to discuss the potential of Roux based methods or the idea of best methods over PM. It'd have to be tomorrow though, its like 1:30 AM here

 

[efattah]

I personally think WaterRoux-EG has good potential for complicated reasons. I took one of Alex Lau's solves and finished it with WaterRoux-EG instead:

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

WaterRoux-EG version:
y x // inspection
M r' F // LF block
R2 x' r U' x // LB pair
R U R' U R2 // solve bottom corners
U2 R' F R2 F' R U2 R' U' R2 // CLL
B U M' U' M B' // pair
M' // setup Waterman L6E
M U r U M U M' U' M' U' r' U' // Waterman L6E
E2 M E2 // permute midges
Total 43 STM

Original Alex Lau Solve
y x // inspection
M r' F // LF block
R2 x' r U' x // LB pair
R' U' R' U2 R' U R2 // RB block
U' R' U r2 R' U' R' // RF pair
U' R' U' R U' R' U R' F R F' U R // CMLL
M' U2 M' U2 M U M' // EO
U M' U2 M' U' M2' // UL/UR
U M' U2 M' // EP
Total 50 STM

 

[efattah]

Another example:

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

WaterRoux EG version:
y2 x // inspection
U R' E2' // LB block
F r' F // LF pair
U' R' U R // bottom corners
F R' F' R' U2 R' U R2 // LEG1
U2 M U' M U2 M2 U' // pair
r U2 R2 U M U' R2 U2 // pair
R // setup L5E
U M' U M' U M U2 M U' R2 // orient midges
U2 M U2 M' // permute midges
48 STM

Original Alex Lau solve:
y2 x // inspection
U R' E2' // LB block
F r' F // LF pair
U2' R U R U2 r' U' R // RB block
U' R U R' U R U r' // RF pair
R U2' R' U2 R' F R2 U R' U' F' // CMLL
M' U M' U M U' M' // EO
U2 M' U2 M' U' // UL/UR
M2' U2 M U2 M' // EP
50 STM

 

[efattah]

3rd example:

F U2 L2 U F2 U D2 F' U2 L2 U' L2 D2 B' L' B2 F2 D L2 U' R' U L2 R U'
WaterRoux-EG version:
y2 // inspection
E R' E' // LF block
R' U R U' B // LB pair
B U2 B' // bottom corners
U R2 U R' U2 R' F' U' F // LEG-1
R U M' U' // 1st redge
M2 R2 U M' U' // 2nd redge
M2 R U M' U2 M U // last pair
R2 // setup L5E
M U' M' U2 M' U' R2 // orient midges
M U2 M U2 M // permute midges
48 STM

Alex Lau version:
y2 // inspection
E R' E' // LF block
R' U R U' B // LB pair
r U' R' U' R U R2 U' R' // RF block
U' M r' U' r // RB pair
U2' F R' F R2 U' R' U' R U R' F2 // CMLL
U2 M' U M' U2 M' U M // EO
U' M U2 M' // UL/UR
U' M' U2 M' // EP
50 STM

 

[crafto22]

@efattah Oh right, I completely forgot about this! This right here has huge potential imo. I'll practice with this for a while and give my thoughts from a speedsolving perspective. @Neuro I'm on vacation atm so PMing would most likely not work out until my return... I might just post some rambling and ideas to my YouTube channel once I get home, spending my nights writing long method proposals is starting to take it's toll.

Sent from my Moto G (4) using Tapatalk

 

[Neuro]

crafto and I are working on a potential way to do edges that we could use some help on in making algs. Basically the DB and FR edges get solved while orienting the rest of the edges so it's possible to end in L5EP. The DB and FR edges get put in the redge slots with no regards to permutation/orientation. Anyone think they can help us out? Preferably keep the algs using M U and R/r moves. I know that the sets with both DB and FR oriented have 16 algs each and there'd be 8 sets so it's not a horribly large algset (120-130 is my guess)

 

[Neuro]

So I'm thinking that the style w/o blockbuilding DR may be better overall. It seems that restricting to a very specific block and influencing the LC would not only be awkward but would take longer than simply putting the last D layer corners in EG/CLL/TCLL+-. Algs may also be shorter because of less edges to perserve. E2L would be really easy because you only need to solve 3 edges rather than 6. Probably same movecount, but it looks like the recog would be much better. Thoughts?

 

[Neuro]

Here's a solve using the WaterRoux-Corners variant, 40 moves!
https://alg.cubing.net/?setup=F2_L-..._M-_U_M_U_M-_U-_M2_U-_M_U2__M_U-//LSE_(14/40)

 

[efattah]

The same example with LMCF L6E finishes in 39:
R F' M' F2 r' z' x'//FB (5/5)
F' U F U L2 D' L' D F2 L2 U l//Corners (12/17)
r U M' U' R' S R2 S' r2//E2L (9/26)
M2 M' U M' U' M U M2 U2 M' U' M2 U2 // LMCF L6E (iDFR set)

A much more ergonomic version without S moves, Waterman style E2L in 41 moves:
R F' M' F2 r' z' x'//FB (5/5)
F' U F U L2 D' L' D F2 L2 U l//Corners (12/17)
r U2 R U' M' U R' U2 R' U M U' // solve 3 edges E2L waterman style
M' U M U' M' U M' U' // L5E DFR
M2 U2 M U2

 

[efattah]

If you didn't know TCLL and only knew LEG-1, two versions:

R F' M' F2 r' z' x'//FB (5/5)
U2 R2 U' R2 // solve red corners
U' R' U R' U' R U L' U R2 U' L // LEG-1 L case
U' R2 U M' U' // 1st edge
R2 U' M U2 M2 U' // 1st pair
r' M // set up
U' M2 U' R2 U M U' R2 U' M' U' // Waterman L6E Set 1 case 1
M U2 M' U2 r // permute midges
50 STM

Another variant:
R F' M' F2 r' z' x'//FB (5/5)
U2 R2 U' R2 // solve red corners
M // centers
U' R' U R' U' R U L' U R2 U' L // LEG-1 L case
U' r M' U M' U' R2 U' M' U // pair
r S' U S U' // pair
U2 M' U M U M' U M U' // L5E
M U2 M U2 M' // permute midges
51 STM

 

[gogozerg]

Original Alex Lau Solve
y x // inspection
M r' F // LF block
R2 x' r U' x // LB pair
R' U' R' U2 R' U R2 // RB block
U' R' U r2 R' U' R' // RF pair
U' R' U' R U' R' U R' F R F' U R // CMLL
M' U2 M' U2 M U M' // EO
U M' U2 M' U' M2' // UL/UR
U M' U2 M' // EP
Total 50 STM

Maybe Alex could have saved a few moves with U'M'U'MUM'U2MU2MUM'U2?
Mhh.. Such an edge orientation pattern may be hard to detect.

 

[Neuro]

Here's a 43 move solve. I'm not really good with E2L hence the length of solution but it had an amazing FB and LSE solution. Not all solves have to end in LSE, this just proves that even without large edge algsets it can still be really efficient.

https://alg.cubing.net/?setup=L2_U2...2_U_R-//E2L M-_U2_M-_U2_M-_U-_M-_U_M2_U-//LSE

 

[efattah]

I finished the same solve above with proper LMCF E2L/L5E and it finished in 39:

Scramble: L2 U2 B2 D2 B' U2 R2 U2 L2 U2 F2 L B R D R2 F U B2 R2 F2
D' R2 y M2 B //FB
U' R' U' R2 F U' F' U R' F2 R U2 R' F R //Corners
U' M2 U M' U' R' B' U' M' U B r2 // E2L
U' M' U' M U M U R' // LMCF L5E BDR set