• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 35,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Edges Last method

Joined
Jul 7, 2008
Messages
627
Location
Saskatchewan, Canada
YouTube
somerandomkidmike
I've named this the "Edges Last Method" because I couldn't think of anything better. I came up with this method a few months ago while messing around with L2Lk. It shares a lot in common with L2Lk/L2L4.

Edit: I've decided to go with TheNextFelix and call the method LC2E. It's not a perfect description, but it does distinguish the method from others.

ANYWAY! Because I don't know how much potential this method has, I've been a little reluctant to share it with the speedsolving community. I DO plan on learning all the algorithms. For the most part, they'd be useful to me, regardless of whether I switch to this method or not.

So, here it is.

Step 1- Left 1x2x3 block (10 moves)

Yup. It's the same as the first step of Roux. If you don't know how to do this, find a different tutorial. I'm not going to explain it.

Step 2- Finish the Left Layer and place 2 Edges on the Second Layer (18 moves)

This is the hardest step, and it requires the most explanation. It is fairly intuitive, but there are several ways you can accomplish this step. Except for a few Advanced cases, you'll probably be using U, R and M moves for this step. In some cases you'll have to move paired blocks "out of the way" before positioning the paired blocks.

Here are just a couple of examples of techniques you can use.

1) Solve the two middle layer edges first, then solve the rest of the first layer.
2) Solve one Edge in the middle layer and 2 pieces in the first layer, then use F2L.
3) Place 2 edges in the middle layer oriented correctly, and position them while solving the first layer.
4) A combination of 1, 2 and 3.

I do plan on providing further documentation to this, but for the most part, I just want to get the general idea out there before I change my mind, and decide not to post anything.

Step 3- CLL (9 moves, 40 cases)
You could use CMLL, CLL, and some Corners First algorithms. The advantage you get here is that there are 2 slots in the middle layer that you can ignore, so you can use the shorter CF algorithms in place of the CLL. (ie. R2 U R U2 R2, rather than R U2 R' U' R U R' U' R U' R')

Step 4- Last 2 Edges (9 moves, 36 cases)
From now on, it's the "edges last" portion. Yes, I got this straight from Stachu Korick's webpage about L2Lk. You can find the algorithms here if you don't know what I'm talking about. http://stachu.cubing.net/l2lk/

Step 5- ELL (11 moves, 29 cases)
ELL has been documented in many different places. This is not new for the speedsolving community. I'm sure you can find information on it.

_____
Total Movecount: 57 STM
Total algorithm count: 105

_____

So, there you have it. Obviously, there are a lot of algorithms to learn for this method, so it might be intimidating. Do I think this method can be any good? I have no clue. I guess I'll wait to see what the rest of you have to say about it.
 
Last edited:

MalusDB

Member
Joined
Mar 23, 2011
Messages
243
Location
Belfast
YouTube
MalusDB91
Step 3- CLL (9 moves, 40 cases)
You could use CMLL, CLL, and some Corners First algorithms. The advantage you get here is that there are 2 slots in the middle layer that you can ignore, so you can use the shorter CF algorithms in place of the CLL. (ie. R2 U R U2 R2, rather than R U2 R' U' R U R' U' R U' R')
I like this! I think I'll give this a whirl :) Thanks for sharing!
 
R

Ramo

Guest
Pretty cool idea. I may try to learn this to take bits and pieces of it and add it on to CFOP!
 
Joined
Jul 7, 2008
Messages
627
Location
Saskatchewan, Canada
YouTube
somerandomkidmike
Can you make an example solve or make a video. I love this idea. My new method.
I will try to sometime today.


Edit:

Scramble: B2 U2 L B2 L' U L F R2 D U R2 D' U R' L' U' D' F' B' R2 D2 R D B

Solution:
x' y

First Block: R2 B r2' U B' U B2 (7 moves)

FL + 2 edges: R u' R' u' R2 u U R2 u2 U R U' R' d' M' U' M (17 moves)

CLL: U' R' U2 R' D' R U2 R' D R2 (10 moves)

L2E: U' L' U L E L' U' L u' (9 moves)

ELL: r U R' U' M U R U' R' U (10 moves)
___
Total moves: 53 moves STM

I hope I got it right. o.o
 
Last edited:

elrog

Member
Joined
Jan 31, 2013
Messages
514
Location
U.S.A.
YouTube
elrog3
I have messed around with alot of different methods that save a portion of the edges for last. If I were to do this method, I'd probably go with getting the cross and two middle edges, then keyholing the 4 corners instead of the first 2 steps. I probably won't learn this method simply because the alg count. I think the EG columns method is a pretty good edges last method that requires a few less algs.
 

Robert-Y

Member
Joined
Jan 21, 2009
Messages
3,287
Location
England
WCA
2009YAUR01
YouTube
Robert271291
I've thought of a different way to do the last 6 edges:

(Assume FL, FR and all U edges need solving)

A. FL or FR + EO
B. EP

Something which is kinda interesting to note is that the solver has "4" options for 'A.'

1: Solve FL and orient edges correctly with respect to R and U
2: Solve FL and orient edges correctly with respect to F and U
3: Solve FR and orient edges correctly with respect to R and U
4: Solve FR and orient edges correctly with respect to F and U
(I'm not sure about the right choice of words, so sorry if you don't understand...)


I haven't managed to calculate how many cases there are for each step, I just wanted to share an idea which I thought was interesting...
 
Last edited:
Joined
Jul 7, 2008
Messages
627
Location
Saskatchewan, Canada
YouTube
somerandomkidmike
I've thought of a different way to do the last 6 edges:

(Assume FL, FR and all U edges need solving)

1. FL + EO
2. EP

I haven't managed to calculate how many cases there are for each step, I just wanted to share an idea which I thought was interesting...
I've thought of that too. I think this type of approach could have many variations that would work.
 

Robert-Y

Member
Joined
Jan 21, 2009
Messages
3,287
Location
England
WCA
2009YAUR01
YouTube
Robert271291
Here's how I would finish that example solve with those other steps:

Scramble: B2 U2 L B2 L' U L F R2 D U R2 D' U R' L' U' D' F' B' R2 D2 R D B

Solution:
x' y

First Block: R2 B r2' U B' U B2 (7 moves)

FL + 2 edges: R u' R' u' R2 u U R2 u2 U R U' R' d' M' U' M (17 moves)

CLL: U' R' U2 R' D' R U2 R' D R2 (10 moves)

y' D2

Step A1: F U R U' R' r U r' U' F'
Step B: U2 R' U' R' U' R' U R U R

Step A2: F' R' U' R' F R U R F
Step B: y' R U' R' U R U R' U2 R2 U R' U' R' U2 R'

Step A3: L u L u' L' U' L' U' L
Step B: U' L U L U L' U' L' U' L'

Step A4: U r U' R' U R r' U' r U r'
Step B: F Hperm F' or y R H perm R'
 
Joined
Jul 7, 2008
Messages
627
Location
Saskatchewan, Canada
YouTube
somerandomkidmike
My average with CF2E is getting close to 20 seconds. If anybody is interested in learning this, or L2Lk, the L2E algorithms are very important.

Anyway, once I learn all the algorithms, and average under 20 seconds, I'll upload an average 3 of 5. The problem I have is that I get REALLY nervous on camera.
 

elrog

Member
Joined
Jan 31, 2013
Messages
514
Location
U.S.A.
YouTube
elrog3
Are you sure this method is unique enoughh to call it a seperate method? It is basically like a Roux varient where you would use CLL-like algs for the 4 right corners rather than the 4 top corners. This is a good way to reduce move count just a little bit, but I'd personally have to get used to the recognition because I'm not color nuetral. As for the last 6 edges, There are many substeps and they are not specific to any method, but can be used with many different methods that reduce the cube to the same position. I think this is just a varient of the Roux method that has a cool and efficient twist to it, a Z' twist to be precise. I do find it interesting that you say it is 57 moves average while the wiki says that standard Roux is 48. The wiki always has the move count slightly low.
 
Joined
Jul 7, 2008
Messages
627
Location
Saskatchewan, Canada
YouTube
somerandomkidmike
Are you sure this method is unique enoughh to call it a seperate method? It is basically like a Roux varient where you would use CLL-like algs for the 4 right corners rather than the 4 top corners. This is a good way to reduce move count just a little bit, but I'd personally have to get used to the recognition because I'm not color nuetral. As for the last 6 edges, There are many substeps and they are not specific to any method, but can be used with many different methods that reduce the cube to the same position. I think this is just a varient of the Roux method that has a cool and efficient twist to it, a Z' twist to be precise. I do find it interesting that you say it is 57 moves average while the wiki says that standard Roux is 48. The wiki always has the move count slightly low.
I personally don't care whether somebody sees this as a separate method. I don't know how you could see this as a Roux variation though. Calling this a Roux variation is like saying Petrus is CFOP variation. I CAN see why you'd say, "It's just a variation of L2Lk".

As far as the movecount goes it's pretty similar to other methods. As I get better at the method, my average movecount seems to be closer to 50-55. However, that has no bearing on how good the method actually is. Some people average over 60 moves with Petrus and CFOP. Some people average over 50 moves with Roux. Some people would easily average over 60 moves with this. A speedsolving method can be considered "good" if people get fast times with it.

I don't care if somebody averages 40 moves or 65 with a method, as long as they can demonstrate that it's fast.
 

elrog

Member
Joined
Jan 31, 2013
Messages
514
Location
U.S.A.
YouTube
elrog3
I'm not sure what L2Lk is, but Ill look into it. It is also true that a good speedsolving method is one that gets good times, but a method gets good times by having 1: good use of inspection time 2: low move count 3: good recognition times and 4: ease in preformance of the moves. The best method for every person varies because different people do each of these things differently. Someone that is good at preforming not so smooth algs may do better with a method that uses less moves and not so smooth algs while somebody that can turn very quickly but only with smooth algs may do well with a method with a higher move count. This brings me to conclude that a method is not necessarily defined as good or bad (unless it is just really bad), and weather a method is good or not depends on who is using it/how it is used.
 
Top