List of 4x4x4 edges methods

[Robert-Y]

Just wondering how many 4x4x4 edges methods there are. So far I know:

Methods:

6-4-2 (hardwick)
6-2-2-2 (hardwick)
Freestyle
Syuhei

2-2-2-2-2-2 variations:

hardwick (again)
avg
mine
Hiroyuke (Take)

I'm guessing there are probably more

 

[AvGalen]

Please replace AVG with 2-2-2-2-2-2. Actually, you could replace all of those n-n-n's with "Hardwick" (I don't think he invented it, but he described it best in the past)

All of these methods are incredibly simular and I wouldn't call them methods but variations.

I would call Caging, K4, Layer-by-Layer and Reduction methods

 

[MTGjumper]

Robert, how is your method is different (meaning, what is your method)?

Also, wouldn't it be more appropriate to call Nakaji's method "freestyle"? I worked out how to use it myself intuitively, and I know others have done the same.

 

[Robert-Y]

Robert, how is your method is different (meaning, what is your method)?

Also, wouldn't it be more appropriate to call Nakaji's method "freestyle"? I worked out how to use it myself intuitively, and I know others have done the same.

Well my method is kinda similar to avg on the 4x4x4 at least (5x5x5 is kinda more different). To summarise what I do, I just put 2 edges next to each other (most of the time it's one in the UF and another one in the LF or RF), and another edge somewhere in the M layer (UB, DF, or DB). Then I slice, replace, slice back to solve 2 edges (on average). I keep doing this until I've finished or I've got exactly 2 edges left. If I do get 2 edges left, I set up and then I use one of these algs:

d2 R F' U R' F d2 or d R F' U R' F d'

I think when you get 2 edges left, it is always possible to set up using 2 moves or less, and do one of the algs above (unless I've been mistaken).

This is only my 4x4x4 edge method. If you try it on the 5x5x5, it's not as good as other methods, which is why I've added more stuff for my 5x5x5 edge method.

 

[James Kobel]

I'm not really sure what method I use so could someone tell me what it is? I set up 1 or 2 edges to be solved then do a Uw to pair them. Then, instead of slicing back, I set up 2 more edges and do Uw until there are at most 4 edges. Then I find the quickest way to pair up the last edges. I'm pretty sure this is called chain pairing but I'm not sure.

 

[AvGalen]

Robert, that is just 2-2-2-2-2-2
James, that is a slight variation of 2-2-2-2-2-2

 

[Lucas Garron]

Please replace AVG with 2-2-2-2-2-2.

I disagree. I did freestyle 2-2-2-2-2-2 for a while, switched to AvG, and got faster. Not looking for that first edge per pair is fast, even if it takes a bit more moves.

 

[EmersonHerrmann]

Syuhei's is 3-3-3-(0, 2, or 3)

 

[Robert-Y]

Robert, that is just 2-2-2-2-2-2
James, that is a slight variation of 2-2-2-2-2-2

What? Are you implying that this variation has been discovered before? If so, I never knew that...

 

[Robert-Y]

Syuhei's is 3-3-3-(0, 2, or 3)

I prefer to reference it as 3-6-(0, 2, or 3) because I think after the first 3, he fixes the centres, then he does a normal 6 edge pairing. (Maybe I'm wrong, but I swear that's what I saw in his tutorial)

 

[AvGalen]

Please replace AVG with 2-2-2-2-2-2.

I disagree. I did freestyle 2-2-2-2-2-2 for a while, switched to AvG, and got faster. Not looking for that first edge per pair is fast, even if it takes a bit more moves.

AvG is exactly the same as 2-2-2-2-2-2 on a 4x4x4. The only difference between freestyle 2-2-2-2-2-2 and AVG 2-2-2-2-2-2 is the use of a fixed "buffer" so you waste a couple of moves, but you have less cube-rotations and better look-ahead.

AvG on a 5x5x5 is exactly the same as on a 4x4x4 only the fixed "buffer" is a semi-pair.

Robert, that is just 2-2-2-2-2-2
James, that is a slight variation of 2-2-2-2-2-2

What? Are you implying that this variation has been discovered before? If so, I never knew that...

Implying that which variation has been discovered before?
Chris Hardwick described edges a long time ago http://www.speedcubing.com/chris/4speedsolve2.html and I don't see any difference between your method and AvG 2-2-2-2-2-2.
I have found James variation before. It uses slightly less moves, but look-ahead is not as good and there are some more problem-cases

 

[Robert-Y]

@Arnaud: The difference between my edge method and yours is what you set up first and last. Maybe you could call my method a variation of yours but I developed mine before I knew about yours. I actually did use to think they were the same but they are slightly different, trust me

 

[fanwuq]

I was just about to make a thread like this!

I'm using 6-2-2-2. I also use freestyle 2-2-2-2-2-2 and 3-6-2 sometimes.

 

[DavidWoner]

I use a variation of 2-2-2-2-2-2 that I discovered on my own. I guess its similar to robert's but we set up in different order. I set up one pair on the M-slice, with UF for the buffer. Then I set up the second pair using [R,L,D,d] move set. Then I pair using a single comm and its mirror (Lw U R U' Rw and Rw U' L U Rw').

example solve(lowercase is wide turn, so r=Rw):

scramble: U F R D B2 Fw F2 Uw' B' Fw' U2 L2 F L' Rw2 F2 Rw' F R2 D' L2 R Uw2 R D' U2 Fw2 L2 R' D' F' R2 Uw B' Fw2 L2 F' L Rw2 B'

opp centers: x2 r2 U' x' U r' x' U r U2 r'

rest of centers: y' F' x' U' r U' r2 / F' x' U r U l2 U2 r / x2 U r U r' l' U2 l

pair 1: x R' F' R' / l U R U' r

pair 2: R B R 3D' / r U' L' U r'

pair 3: L' U' R' x / l U R U' r

pair 4: x' R U R' 3D' / r U' L' U r'

9th edge: L' U' B L / r U' L' U r'

last 3: 3D' / l U R U' r

I ran through that like 5 times so I hope its all correct. I avg 20-25 seconds for edges with this method.

 

[AvGalen]

@Arnaud: The difference between my edge method and yours is what you set up first and last. Maybe you could call my method a variation of yours but I developed mine before I knew about yours. I actually did use to think they were the same but they are slightly different, trust me

I don't understand why you say "trust me". Why not explain it?

How would you solve this scramble?

Scramble (57 moves in total, but a wrote them in an easy to follow manner. small letters mean wide turns)
r U r' U r U2 r'
F' U
l' U2 l U l' U l
L2 R2 F2 B2

r U r' U r U2 r'
F' U
l' U2 l U l' U l
L R' F2 D' B2

r U r' U r U2 r'
F' U
l' U2 l U l' U l

I would do (my buffer is uLF, so I pair on the front E-slice.):
B2 R2 B', d R U R' d'
B R2, d R' D' R d'
R' B', d F' U2 F d'
R2 B', d F' U2 F d'
F D' z', d F D' F' d'
R, d R U' R' F' U2 F d'

For 6-4-2 it would be
B2 R2 y, R2 y, SKIP!!!!, d
R U2 R' y', R F' R' F y', F' U' F, d'
B R2 y, F D2 F', d
F' U2 F y', U R U' R', d'
R y, d2 R U' R' F' U2 F d2

Vault312, I tried you example, but centers don't seem to work. The two opposites work, then it goes bad. What do you mean with small letter turns?

 

[cubacca1972]

I don't know how to classify my edge solve method, but it goes like this:

After solving two opposite centers and solving the corners, build edge pairs one at a time and direct solve them until two opposite sides are solved (L and R).

Next, sort the l edge cubies and r edge cubies, so that they are in their appropriate layer (1 of 7 algorithms).

Finally, solve all 4 remaining edge pairs (1 of 8 algorithms). Usually, an l or r layer move will be required to align the edge pairs.

You can see the algorithms here:

http://cubacca1972.googlepages.com/alternatephase2:solvetheremainingedgecub

 

[Robert-Y]

Ah I feel stupid lol. I thought I understood avg, NOW I think I understand avg and now I think our variations aren't as similar as I thought they were. I was talking to Dan Cohen a few minutes ago, he basically uses something similar to me. My method involves many cube rotations. I'll post a solution to how I would do your scramble in a couple of minutes. I'm gonna eat dinner now.

 

[fanwuq]

Arnaud's scramble, my method:

U2/ U'L'U/ x2LUL'U'/ l' 9
R'FRF'/ xU'R2U/ xU'R'U/ r 20
D/ l'/ UL2U'/ r 26
L'U'/ r/ FR2F'/ r' 33
U'L/ x'dRF'UR'Fu' 42

6-2-2-2 M-slice.
After watching arnaud's solution, I'm liking 6-4-2. One thing I hate about 6 pairing is when there are 3- cycles, it really sucks. It takes me a while to realize it when I see that situation. I usually continue and solve 5 edges instead. But now I think I realize if you do a 2 pair method, you end up solving all 3 edges. So now, when I see it, I should solve as 2-6-2. That would actually skip a step.


my solve replayed on Lucas's really cool applet

 

[AvGalen]

cubacca1972: I would call that a different method. What times do you get with it?
Robert-Y: I will look at your solutions tomorrow
fanwuq: 6-4-2 can be really nice (like this example) or horrible. On average 2-2-2-2-2-2 is faster and much more consistent.

I hope everyone uses my scramble if they want to compare their edge-pairing method. The scramble keeps centers solved and can be performed really fast.

 

[Robert-Y]

Ok I'll go through it in good detail

R=Red
O=Orange
Y=Yellow
G=Green
B=Blue
W=White

(Scramble with white on top and green at the front)

1st pair: I can see the red and yellow pair in front of me so I do z' R U to set up the RYs. The RY in the LFU has to replace the OY in the UFL so I need to find the other YO edge. It's in the BL slot, so I simply do B to place it correctly in the M layer. Then I simply use slice, replace, and slice back: r2' (slice), U L' U' (replace), r2 (slice back), to fix RYs and OYs.

2nd pair: Searching around, I can see the GYs, so I set them up by doing x U'. The GY in the LFU has to replace the GO in the UFL so I try to find the other GO edge which is in the RF slot. To place it in the M layer, I do R2 B. Then I fix the GYs and GOs by doing r' U L' U' r.

3rd pair: Immediately after the 2nd pair of edges, I can see the OW edges and to set them up, I just do U. The OW in the RFU has to replace the RW in the UFR. The other RW is in the RD slot. To place it in the M layer, I do D'. To fix the OWs and RWs, I do l' U' R U l.

4th pair: BWs. Set up: x R'. BW in RFU must replace BR in UFR. Other BR is in LF, so to place in M layer: L D. To fix BWs and BRs: l' U' R U l.

5th pair: BYs. Set up: x z' U'. BY in UFL must replace BO in UFL. Other BO is in UL, so to place in M layer: L' B' L. To fix BYs and BOs: r2' U L' U' r2.

6th pair (last two). GRs and GWs. y R to set up. d R F' U R' F d' to fix.

This is a basic example and doesn't cover "special avg like cases" and the problem(s) where you set up two edges which contain the same coloured edges e.g. two edges with BO and BR.

Hope that helped, (sorry if there are mistakes anywhere.)