Robert's 4x4x4 edge-pairing method examples

[Robert-Y]

This was requested by minigoings and I thought it would be a bit more useful than my video tutorial. In this thread, I'll try and take you through some example solves, showing you my thoughts as I go along. This will be a lot like "Arnaud's 5x5x5 Edge-Pairing method: Examples" thread


Here's how I would basically solve the edges (after finishing the centres):

Some key terms to remember before reading:

Dedge: A dedge is a pair of edges next to each other.
M layer: The l and r slice.
Wide turn: A double or triple layer turn.


1a. Find two edges which go together. Let's call them "A" edges.
1b. Put the two "A" edges next to each other so that they're either in the LFu and UFr spots, or RFu and UFl spots.

2a. Look at the edge next to the A edge in the UFl or UFr spot. Let's call it a "B" edge.
2b. Find the other "B" edge and without disturbing the positions of the "A" edges and the "B" edge in the UF dedge, place it in the M layer so that one B edge is in the l slice and the other one is in the r slice. So when you do a wide turn, the "B" edges will be paired up.

3a. Pair up the "B" edges by doing a wide turn so now you have a solved dedge.
3b. Replace the solved B dedge with the LF or RF dedge containing an "A" edge by doing either U L' U' (for LF dedge) or U' R U (for RF dedge)
3c. Undo the wide turn to restore the centres. Now 2 dedges have been solved.

During step 3, you should looking ahead for your next pair of "A" edges and maybe even "B" edges

Now you keep repeating these steps over and over until you're done or you're left with two unsolved dedges. (If you are left with two unsolved dedges, you can simply place them both in the E layer, so that the edges are mirrored vertically. Then do: Dw R F' U R' F Dw' to solve them)


There are of course problems with this method (as with any other method...). These problems will hopefully come up and I'll show you how to deal with them in the following examples.

 

[Robert-Y]

Now let's do some examples.

(To follow the following examples p) you will need to have a cube with the western colour scheme (opposites are: white-yellow, green-blue, red-orange). Also, you must scramble with white on top and green on front).

Note:
1. "a" refers to the setting up of the "A" edges.
2. "b" refers to the setting up of the "B" edges OR placing an unsolved dedge in the UB, BD or FD spot.
3. "c" refers to the the turns you execute pair up the edges and possibly why you do those turns.



1st example:


Scramble:

Rw U Rw'
U D'
Rw U Rw'
F2 B2
Rw U Rw'
R2 B'
Rw U Rw'
U' Lw' U Lw U
Rw U Rw'


Solution:

(the red-whites immediately stood out for me)

1.
a: red-whites: x y2
b: green-yellow: B
c: solve the A and B edges: Rw' (U L' U') Rw

(whilst solving them, I spotted the green-oranges)

2.
a: green-orange: L U
b: green-red: R B
c: solve the A and B edges: Rw' (U L' U') Rw

(just before I solved them, I noticed the blue-yellows standing out at me)

3.
a: blue-yellow: y R'
b: blue-orange: B
c: solve the A and B edges: Lw (U' R U) Lw'

(just before I solved them, I noticed the green-whites on the F face)

4.
a: green-white: U2
b: blue-red: D'
c: solve the A and B edges: Rw2' (U L' U') Rw2

(whilst solving them, I spotted the yellow-oranges and yellow reds, and the fact that there's a small problem)

5.
a: yellow-orange: x U'
b: place an unsolved dedge in the UB spot: (fortunately it's been done)
c: solve the A edges and immediately set up the B edges (UFr and RFd) so they become the next A edges: Lw (U' R U) Lw' x' R'

(I notice that I probably have 3 dedges left and that the other B edge is somewhere in the BD spot)

6.
a: yellow-red: already done
b: white-orange: B2
c: solve the A and B edges (as well as the other 2 edges): Lw (U' R U) Lw'

Done!



2nd example:(a bit more advanced)


Scramble:

Rw U Rw2' F Rw
L2 R2 B2
Rw U' Rw2' F' Rw


Solution:

(I somehow spotted the blue-yellows on the left and that by doing just U', I will have set up the A and B edges)

1
a: blue-yellow: U'
b: red-white: (already done)
c: solve the A and B edges: Rw' (U L' U') Rw

(Just before the Rw, I noticed the orange-whites)

2
a: orange-white: x2' y U'
b: green-orange: R2 B
c: solve the A and B edges: Rw' (U L' U') Rw

(As I was about to set up the green-oranges, I spotted the green-reds and decided to keep an eye on them whilst doing Rw' (U L' U') Rw)

3
a: green-red: R2'
b: green-yellow: R B R' (one of the longest cases but at least you can spot it at once)
c: solve the A and B edges: Lw (U' R U) Lw'

(After the Lw', I saw the red-yellows on the right)

4
a: red-yellow: U
b: blue-orange: B2
c: solve the A and B edges: Lw (U' R U) Lw'

(After the Lw', I noticed the green-whites, and the blue-whites next to them (i.e. we have a problem))

5
a: green-white: y' U2
b: place an unsolved dedge in the FD spot: (fortunately it's been done)
c: solve the A edges and immediately set up the B edges (UFr and RFd) so they become the next A edges: Lw' (U' R U) Lw x R

(After setting up the blue-whites, I quickly searched and found the other B edge at the RD spot)

6
a: blue-white: (already done)
b: yellow orange: D
c: solve the A and B edges (as well as the other 2 edges): Lw2 (U' R U) Lw2'

Done!

 

[dunpeal2064]

This is so helpful. Thank you Rob! I love big cubes but I am terrible at the edge pairing. takes up 1:30-2:00 of my solves. This will help very much

 

[miniGOINGS]

Awesome!!!

 

[Toad]

I also use Rob's method so if anyone has any questions you can ask me too... But this tutorial looks pretty conclusive haha

EDIT: I'm not as dire at 4x4 as my WCA profile suggests... Stupid Eastsheen back then

 

[sz35]

Awesome tutorial! I don't like this method, I prefer the nakaji pairing, but still V GOOD!

 

[Jude]

Wow this is how I used to solve 4x4x4 edges when I first figured it out (and I actually still do for 5x5x5 last 4 edges in some cases) . I had no idea this method had any potential. The only difference is I would set up the 'b' edges first then have the 2nd 'a' edge in either BR or BL, so to solve I would do Lw' U R' U' Lw (or it's reflection ). To see how I'd set it up just do Lw' U R U' Lw on a solved cube.

 

[Robert-Y]

Update: Added another example which is a bit more advanced

Awesome tutorial! I don't like this method, I prefer the nakaji pairing, but still V GOOD!

Thanks, and btw the only people who I know can average under 50 seconds with "Nakaji's" method is still just Yu Nakajima himself. It seems to me that none of us apart from Yu knows how to get very fast with his method

Wow this is how I used to solve 4x4x4 edges when I first figured it out (and I actually still do for 5x5x5 last 4 edges in some cases) . I had no idea this method had any potential. The only difference is I would set up the 'b' edges first then have the 2nd 'a' edge in either BR or BL, so to solve I would do Lw' U R' U' Lw (or it's reflection ). To see how I'd set it up just do Lw' U R U' Lw on a solved cube.

I used to do something like this and I also used to think that AVG was just like my method except the B edges are set up first then the A edges.

 

[Escher]

Awesome tutorial! I don't like this method, I prefer the nakaji pairing, but still V GOOD!

Thanks, and btw the only people who I know can average under 50 seconds with "Nakaji's" method is still just Yu Nakajima himself. It seems to me that none of us apart from Yu knows how to get very fast with his method

Hey now! I average between 30-35 for redux, if I had a decent 3x3 stage that would be sub 50 average

 

[MTGjumper]

I also average 30-35 for redux with Nakajima's method, but my 4x4 isn't very good for 3x3

 

[Robert-Y]

If you and Rowan want to be sub-50, then it might be easier to improve on your reduction, I think I can average about 25 seconds for reduction and about 15-20 seconds for the 3x3x3 stage (that's including parity)

 

[Escher]

If you and Rowan want to be sub-50, then it might be easier to improve on your reduction, I think I can average about 25 seconds for reduction and about 15-20 seconds for the 3x3x3 stage (that's including parity)

To be completely honest I think we just need to improve tps
Given that other methods use about 30 moves for edge pairing and Nakaji's uses about 60...

 

[masterofthebass]

If you and Rowan want to be sub-50, then it might be easier to improve on your reduction, I think I can average about 25 seconds for reduction and about 15-20 seconds for the 3x3x3 stage (that's including parity)

15-20 toooooo fast.

 

[Hyprul 9-ty2]

Must. Get. That. SUB1. AVG.

 

[Robert-Y]

If you and Rowan want to be sub-50, then it might be easier to improve on your reduction, I think I can average about 25 seconds for reduction and about 15-20 seconds for the 3x3x3 stage (that's including parity)

15-20 toooooo fast.

Meh, I know Syuhei, Haixu, and probably Han Cyun, can average about the same for the 3x3x3 stage. As for you... I don't know, you haven't switched back to reduction-3x3x3 have you?

 

[masterofthebass]

If you and Rowan want to be sub-50, then it might be easier to improve on your reduction, I think I can average about 25 seconds for reduction and about 15-20 seconds for the 3x3x3 stage (that's including parity)

15-20 toooooo fast.

Meh, I know Syuhei, Haixu, and probably Han Cyun, can average about the same for the 3x3x3 stage. As for you... I don't know, you haven't switched back to reduction-3x3x3 have you?

Well since with parity my LL takes 10s or more, I doubt I get 15-20 3x3 even with the damn cross already done.

 

[HASH-CUBE]

Hi Robert, your tutorial is very awesome! i tried it and my times got better
hopefully i will try to improve myself using this method and thanks for it

 

[HASH-CUBE]

I made an explanation video:

 

[SebCube]

Yaaay this is awesome, this helped so much, thank you.

 

[negative_earth]

i used to solve 4x4x4 in 2-3 minutes...

this method reduces 30-50 seconds of my solving time!

thank yooouuuuuu ^_^v

btw, i solve centers in 35-50 seconds

is there any way to get faster in solving the centers?