New 4x4x4~NxNxN Method: NJPB Method

[JackTriton]

Hi, I'm Jack Triton (former Bicube single ao5 UWR, current ao12 UWR holder).
Today, I'm going to present my newest method called NJPB Method (NxNxN Jack('s) Pair & Block).
This method is mixture of Yau and APB, but it's much faster than it sounds.

Pros compared with Yau are:

• Less dynamic movement(x, y, z)
• When you redact it to 3x3x3 state, the 223 (334) block is created on your dL
• Easier recognition for edge pairing
• No new algorithms to remember

And this is how to do it:

1. F2C (First 2 Center)
   Solve 2 opposite centers on L and R. (Center on L is should not be your cross color, if you're not Color Neutral)
   
2. 3CEBC (3 Cross Edges Bottom half of Centers pairing)
   Solve 3 of the 4 cross dedges like Yau and place on L with pairing those with the correct bottom half of centers like the image. (Avoid placing on U layer and orient correctly.)
   The one on the D should be cross color, if you're not color neutral.
   
3. D Block Insert
   Insert correct paired DFL and DBL Corner-Half edge pairs.
   You can do this step with 3CEBC at the same time.
   You can also do the exact thing by solving second layer after first step, then insert edges to the slot which is something similar to F2L or first block.
   
   
   
4. RCdE (Rest Center & d Edge insertion)
   Solve rest of centers with pairing the bottom edges with correct edge using only U, Rw for mainly and outer layers for inserting.
   
5. L7E (Last 7 Edges)
   Solve the last 7 edges (Colored Edges) using only U, Rw for mainly and outer layers for inserting.
   I recommend using 2-2-3, 3-3 or 2-1-3 pairing.
   For advanced, solve EO as well using F moves.

After this, The 4x4x4 will be the same state as APB's 223 Block paired.
So doing the same thing as you solve this to 3x3x3 + OLL Parity and PLL Parity would be the last step.

I personally got 1 minute faster compared with using Yau.
Example)
Scramble: U' L2 U2 F2 U2 F' R2 F' L2 B' D2 B L U F' B' L B2 L D2 Uw2 R Rw2 U Uw2 B2 R Uw2 L' U2 L' Fw D L' U' F' D Fw D2 Rw U' Uw2 F' Fw2

Solution:
Rw' U' Rw U L' Rw F Rw' U' Fw' R Fw2 L' Fw U' R Fw Lw F2 U Rw B2 Bw' R' Bw // F2C + 3BC + 1CE
Lw2 F Lw2 R F U F' L' B L B' R U' R' U Rw' U2 B // 3CEBC
U2 R U' R' U' F2 U F2 R' U R U B2 U' B2 // D Block Insert
Rw U Rw2 U' Rw' U' Rw' U2 Rw' U' R' U2 R U' Rw2 // RCdE
U' R2 U Rw' R2 F R F' Rw U F' U F U' Rw U R' U' Rw' // L7E (3-3)
R2 U (Rw2 B2 Rw' U2 Rw' U2 x' U2 Rw' U2 Rw U2 Rw' U2 Rw2 U2 x) // EO + OLL Parity
R' U' R U' R U R' U' R2 U R2 U R' U2 R U R' U // DBR 2x2x1 block +DFR Pairing
U R U R' U' L' U2 R U R' U2 L // WVLS-13
(x' R U' R' D R U R' D' R U R' D R U' R' D' x) U2 // E-perm, finish

Thanks for reading!

 

[NebraskanSpeedsolver]

Hi, I'm Jack Triton (former Bicube single ao5 UWR, current ao12 UWR holder).
Today, I'm going to present my newest method called NJPB Method (NxNxN Jack('s) Pair & Block).
This method is mixture of Yau and APB, but it's much faster than it sounds.

Pros compared with Yau are:

• Less dynamic movement(x, y, z)
• When you redact it to 3x3x3 state, the 223 (334) block is created on your dL
• Easier recognition for edge pairing
• No new algorithms to remember

And this is how to do it:

1. F2C (First 2 Center)
   Solve 2 opposite centers on L and R. (Center on L is should not be your cross color, if you're not Color Neutral)
   View attachment 29048
2. 3CEBC (3 Cross Edges Bottom half of Centers pairing)
   Solve 3 of the 4 cross dedges like Yau and place on L with pairing those with the correct bottom half of centers like the image. (Avoid placing on U layer and orient correctly.)
   View attachment 29049
3. D Block Insert
   Insert correct paired DFL and DBL Corner-Half edge pairs.
   You can do this step with 3CEBC at the same time.
   View attachment 29050
4. RCdE (Rest Center & d Edge insertion)
   Solve rest of centers with pairing the bottom edges with correct edge using only U, Rw for mainly and outer layers for inserting.
   View attachment 29051
5. L7E (Last 7 Edges)
   Solve the last 7 edges (Colored Edges) using only U, Rw for mainly and outer layers for inserting.
   I recommend using 2-2-3, 3-3 or 2-1-3 pairing.View attachment 29052
   For advanced, solve EO as well using F moves.

After this, The 4x4x4 will be the same state as APB's 223 Block paired.
So doing the same thing as you solve this to 3x3x3 + OLL Parity and PLL Parity would be the last step.

I personally got 1 minute faster compared with using Yau.
Example)
Scramble: U' L2 U2 F2 U2 F' R2 F' L2 B' D2 B L U F' B' L B2 L D2 Uw2 R Rw2 U Uw2 B2 R Uw2 L' U2 L' Fw D L' U' F' D Fw D2 Rw U' Uw2 F' Fw2
View attachment 29053
Solution:
Rw' U' Rw U L' Rw F Rw' U' Fw' R Fw2 L' Fw U' R Fw Lw F2 U Rw B2 Bw' R' Bw // F2C + 3BC + 1CE
Lw2 F Lw2 R F U F' L' B L B' R U' R' U Rw' U2 B // 3CEBC
U2 R U' R' U' F2 U F2 R' U R U B2 U' B2 // D Block Insert
Rw U Rw2 U' Rw' U' Rw' U2 Rw' U' R' U2 R U' Rw2 // RCdE
U' R2 U Rw' R2 F R F' Rw U F' U F U' Rw U R' U' Rw' // L7E (3-3)
R2 U (Rw2 B2 Rw' U2 Rw' U2 x' U2 Rw' U2 Rw U2 Rw' U2 Rw2 U2 x) // EO + OLL Parity
R' U' R U' R U R' U' R2 U R2 U R' U2 R U R' U // DBR 2x2x1 block +DFR Pairing
U R U R' U' L' U2 R U R' U2 L // WVLS-13
(x' R U' R' D R U R' D' R U R' D R U' R' D' x) U2 // E-perm, finish

Thanks for reading!

I'll spend some time and try this out this weekend.

Looks really cool!

 

[1x1 wr holder!]

So this is like that 3x3 block method or bar method or whatever (the one where you make a 2x2x3 in the bottom left and then solve with R/U layers). Wow, this could be big. I can see this easily decreasing the number of cube rotations while also making look ahead easier. One minute faster huh? Wowza, I have homework, but who cares? Thanks Mr. Triton.

 

[Christopher Mowla]

View attachment 29051

R2 U (Rw2 B2 Rw' U2 Rw' U2 x' U2 Rw' U2 Rw U2 Rw' U2 Rw2 U2 x) // EO + OLL Parity

You could use faster Petrus Parity algs.

For example, Rw' U R' U2 R U' Rw' U2 Rw' U2 Rw' U' R U2 R' U Rw' or Rw' U' L' U2 L U Rw' U2 Rw' U2 Rw' U L' U2 L U' Rw'.

 

[Filipe Teixeira]

Or you could just use petrus

 

[xyzzy]

2nd step is something I've been doing since ~2016 (ex.). Until early 2017, the way I continued from there was to solve four F2L pairs (each consisting of one corner and one wing), CLL, then continue as in Sandwich; then I switched to just finishing L4C and continuing as in Yau.

(… except with a bunch of other changes as well, effectively ship-of-Theseus-ing out Rob Yau's contributions to the Yau method, but that's irrelevant to the current topic)

The next few steps are pretty much the same as in OBLBL.

(Iirc if you try to extend OBLBL to 6×6×6 and up, obliques become annoying to deal with once you're done with the left half of the puzzle.)

You could use faster Petrus Parity algs.

For example, Rw' U R' U2 R U' Rw' U2 Rw' U2 Rw' U' R U2 R' U Rw' or Rw' U' L' U2 L U Rw' U2 Rw' U2 Rw' U L' U2 L U' Rw'.

This variant of the first alg might be promising. I can execute it a bit faster than my usual OLL parity algs. (Regrip spam during the r U2 r U2 r, I guess. Lots of left-hand overworking because your right hand isn't available to do the U2s.)

 Twizzle link 

r U' R U2 R' U r U2 r U2 r U R' U2 R U' r

There is the benefit that these flip 3 edges, so even if they're a bit slower in isolation, they might still be faster overall compared to F R* F' + OLL parity. However, these algs also have the downside that they have a huge "blast radius" – changing so much of the cube state also makes it harder to look ahead to the next step.

 

[JackTriton]

Or you could just use petrus

Petrus would restrict some moves, but this would not have effect to the solve itself: mostly.

The next few steps are pretty much the same as in OBLBL.

I looked up, and yes it is almost identical.
Looked up who made this and I realized it's the same person who had the former world record of Bicube (before I got).
What a coincidence!
But still, this method would be revolutional for the Yau users.

(Iirc if you try to extend OBLBL to 6×6×6 and up, obliques become annoying to deal with once you're done with the left half of the puzzle.)

If you do slot pairing for the specific layers, it would be much easier to pair the centers.
Think each connected colors as paired, below showing are the example pairing 1 step before the finish.
6x6x6 (Use mostly 3Rw, U and Rw),

7x7x7 (3Rw, U and Rw),

8x8x8 from left (4Rw, U, 3Rw and Rw) to right (3Rw, U and Rw).

Then, do the same pairing as 5x5x5 would be much faster and can be applied on pretty much every higher NxNxNs.
Edge pairing can be done on L7E phase, or skip some of them while pairing centers.

But anyway, thanks for the review!

 

[Jorian Meeuse]

Looks interesting, I'll definitely try it out!

View attachment 29065
View attachment 29066
View attachment 29067View attachment 29068

BTW that is a very scary colour scheme

 

[randomness...]

i really like this for 5, i just need to get comfortable in untimed solves, and learn more into it
I will definitely put some time into this, because my tps is high in these (r,R,u,U)

 

[JackTriton]

Update:

Since step 4 and step 5 have same rotate dimensions, you can do both of them at the same time (at least 3 or 4 edges are easy to solve).
For the tips, move all edges that is already solved during step 4 to other than UF and UB would be great and will be the time saver.

 

[JackTriton]

Update 2:
For the edge pairing for NxNxN greater than 6, you can do both expand center building (which explained on #7) and edge pairing at the same time (I call this pairing step as "Expand Pairing" or ExP for short).

This allows you to get non-complex 5x5x5 L7E at the last edge pairing after solving the RCdE, which means you can bring this to 3x3x3 state much faster than original edge pairings.