Noahaha
blindmod
As some of you may remember, I proposed a Petrus based CP-block method a while back. The problem with it was that it made Petrus even more inefficient than it already was just in order to have a 1LLL.
I never dropped the idea of CP-blocks, however...
...and today I bring you my NEW AND IMPROVED CP-block method. Unlike the last one, it is its own method. It shares characteristics with Petrus, Roux and OBLBL, but the end result is definitely something entirely different and hopefully new. I say hopefully, because all "new" methods must come with the disclaimer that someone may have discovered them before.
Onto the method:
Step 1: CP (Roux) Block
The goal of this step is to reach a state where there is a Roux block solved on the left and the corners are permuted.
It can be divided into three parts:
1a - SOLVE A ROUX BLOCK on the left (note that you can mirror the entire method) just like the first step of Roux. I recommend doing the same block every time.
1b - PLACE, but do NOT orient the FDR and BDR corners. This should take 4 moves at most.
1c - CP.
Recognition: figure out which two U-layer corners need to be swapped.
If no corners need to be swapped, you're done!
If two adjacent corners need to be swapped, place them at UFR and UBR, and do F' U F or y R U' R' y'.
If two diagonal corners need to be swapped, make them adjacent and then do what's above.
Cases:
- UFL UFR = F' U' F
- UFR UBR = F' U F
- UBR UBL = F' U2 F
- UBR UFL = U' F' U' F
- UFL UBR = R' F' U' F
- UBL UFR = R F' U2 F
Ideally, 1a and 1b would both be planned in inspection, or at the very least the positions of the last two D-layer corners would be noted.
Step 2: EDGES!
This step has two parts, although ideally they should merge into one step. Step 2 is where this method has a LOT of freedom.
Step 2 brings the cube to a state where the bottom left 2x2x3 is solved, the edges are oriented and the corners are permuted. As long as you only use U, R and r moves during this step, the corners will remain permuted from the CP-block.
2a - ORIENT ALL EDGES
I haven't found a way to break this down yet, so people who are not familiar with Roux might struggle a little bit here. A Roux approach seems best to me, using moves like M' U M to orient 4 edges at a time. If you have two edges left, just put them at UB and UL and do M' U M U2 M' U M. Note that an M/M' move changes the orientation of all edges on the M-slice. You should start by placing your U/D centers into place or opposite places because in the end you have to have them oriented along with the edges.
As qq pointed out, there are 5-move 2-flips:
DR and UF: r U R U' r'
FR and UF: M' U R U' r'
etc. they all follow the trend of slice, replace an edge, slice back.
2b - FINISH THE 2x2x3 by placing the DF and DB edges.
This is extremely easy and should only require 3-6 moves depending on where your edges are at the end of 2a. A good strategy is to connect them at UL and UR, and then insert them between the correct two centers. Remember, you can only do double turns involving the M-slice in order to preserve EO, so other than M U2 M' type things, the moves you can make are: R, R2, R', M2 and r2.
NOTE: you can start step 3 before finishing 2a if it's convenient.
Step 3: Right Block
This is the same as the right block in ZZ and step 4 of Petrus. Just use R and U to finish F2L.
Step 4: 2GLL
Yay! We have finally reached our 1LLL, and it only requires 85 algs. That's not much more than CFOP
Summary
Step 1 - CP block
-1a = Roux Block
-1b = Place last two D-layer corners
-1c = CP in 0, 3, or 4 moves
Step 2 - EDGES
-2a = EO
-2b = Finish Petrus Block
Step 3 - Right Block
Step 4 - 2GLL
Analysis
Pros:
- REALLY fingertrick friendly (only uses M, U, R and r after step 1)
- Practically rotationless
- Many substeps can be solved in very few moves
- 1LLL
- Lots of freedom
- It's really fun!!!
Cons:
- A lot needs to be planned in inspection
- Lots of sub-steps
- Potentially really high movecount
- 85 algs to learn
- Although there are often really quick solutions, they are not always easy to see in a speedsolve.
Feel free to give me more pros and cons to add
Example Solve
Scramble (in solving orientation): B2 U2 L2 R2 D' R2 D' B2 D' B2 U2 L B F2 L2 D' L' F2 R' D F2
1a: U2 L' U L2 D F U' L2 (8)
1b: U' R' U R (4)
1c: U2 B U' B' (4)
2a: R' U2 M' U M (5)
2b: R2 U' r' U2 r (5)
3: R U2 R' U2 R' U R U2 R' U R (11)
4: U2 R' U R' U' R' U' R' U R U R2 (12)
Move count: 49
Conclusion
I don't think anyone will end up using this to speedsolve because it is pretty complicated, but I hope people enjoy the method and use it for fun sometimes. If you like it, please post an example solve. A lot of the fun of this method is finding shortcuts and ways to accomplish two things at once.
Sorry for the long read, and sorry if this has been thought of already!
I never dropped the idea of CP-blocks, however...
...and today I bring you my NEW AND IMPROVED CP-block method. Unlike the last one, it is its own method. It shares characteristics with Petrus, Roux and OBLBL, but the end result is definitely something entirely different and hopefully new. I say hopefully, because all "new" methods must come with the disclaimer that someone may have discovered them before.
Onto the method:
Step 1: CP (Roux) Block
The goal of this step is to reach a state where there is a Roux block solved on the left and the corners are permuted.
It can be divided into three parts:
1a - SOLVE A ROUX BLOCK on the left (note that you can mirror the entire method) just like the first step of Roux. I recommend doing the same block every time.
1b - PLACE, but do NOT orient the FDR and BDR corners. This should take 4 moves at most.
1c - CP.
Recognition: figure out which two U-layer corners need to be swapped.
If no corners need to be swapped, you're done!
If two adjacent corners need to be swapped, place them at UFR and UBR, and do F' U F or y R U' R' y'.
If two diagonal corners need to be swapped, make them adjacent and then do what's above.
Cases:
- UFL UFR = F' U' F
- UFR UBR = F' U F
- UBR UBL = F' U2 F
- UBR UFL = U' F' U' F
- UFL UBR = R' F' U' F
- UBL UFR = R F' U2 F
Ideally, 1a and 1b would both be planned in inspection, or at the very least the positions of the last two D-layer corners would be noted.
Step 2: EDGES!
This step has two parts, although ideally they should merge into one step. Step 2 is where this method has a LOT of freedom.
Step 2 brings the cube to a state where the bottom left 2x2x3 is solved, the edges are oriented and the corners are permuted. As long as you only use U, R and r moves during this step, the corners will remain permuted from the CP-block.
2a - ORIENT ALL EDGES
I haven't found a way to break this down yet, so people who are not familiar with Roux might struggle a little bit here. A Roux approach seems best to me, using moves like M' U M to orient 4 edges at a time. If you have two edges left, just put them at UB and UL and do M' U M U2 M' U M. Note that an M/M' move changes the orientation of all edges on the M-slice. You should start by placing your U/D centers into place or opposite places because in the end you have to have them oriented along with the edges.
As qq pointed out, there are 5-move 2-flips:
DR and UF: r U R U' r'
FR and UF: M' U R U' r'
etc. they all follow the trend of slice, replace an edge, slice back.
2b - FINISH THE 2x2x3 by placing the DF and DB edges.
This is extremely easy and should only require 3-6 moves depending on where your edges are at the end of 2a. A good strategy is to connect them at UL and UR, and then insert them between the correct two centers. Remember, you can only do double turns involving the M-slice in order to preserve EO, so other than M U2 M' type things, the moves you can make are: R, R2, R', M2 and r2.
NOTE: you can start step 3 before finishing 2a if it's convenient.
Step 3: Right Block
This is the same as the right block in ZZ and step 4 of Petrus. Just use R and U to finish F2L.
Step 4: 2GLL
Yay! We have finally reached our 1LLL, and it only requires 85 algs. That's not much more than CFOP
Summary
Step 1 - CP block
-1a = Roux Block
-1b = Place last two D-layer corners
-1c = CP in 0, 3, or 4 moves
Step 2 - EDGES
-2a = EO
-2b = Finish Petrus Block
Step 3 - Right Block
Step 4 - 2GLL
Analysis
Pros:
- REALLY fingertrick friendly (only uses M, U, R and r after step 1)
- Practically rotationless
- Many substeps can be solved in very few moves
- 1LLL
- Lots of freedom
- It's really fun!!!
Cons:
- A lot needs to be planned in inspection
- Lots of sub-steps
- Potentially really high movecount
- 85 algs to learn
- Although there are often really quick solutions, they are not always easy to see in a speedsolve.
Feel free to give me more pros and cons to add
Example Solve
Scramble (in solving orientation): B2 U2 L2 R2 D' R2 D' B2 D' B2 U2 L B F2 L2 D' L' F2 R' D F2
1a: U2 L' U L2 D F U' L2 (8)
1b: U' R' U R (4)
1c: U2 B U' B' (4)
2a: R' U2 M' U M (5)
2b: R2 U' r' U2 r (5)
3: R U2 R' U2 R' U R U2 R' U R (11)
4: U2 R' U R' U' R' U' R' U R U R2 (12)
Move count: 49
Conclusion
I don't think anyone will end up using this to speedsolve because it is pretty complicated, but I hope people enjoy the method and use it for fun sometimes. If you like it, please post an example solve. A lot of the fun of this method is finding shortcuts and ways to accomplish two things at once.
Sorry for the long read, and sorry if this has been thought of already!
Last edited:
I'm planning on releasing it someday, but it is not quite ready yet as if now. I'd be much more interested in other people looking into it as well, with unspoiled eyes xD
