The New Method / Substep / Concept Idea Thread

[FinnTheCuber]

My idea is to start with the first blocks on roux. Then continue to sole the other two cross edges. Then proceed with normal oll and ppl. Dont know if this is new or not. And if it is, could someone please test it for me?

 

[ProStar]

This is proposed a lot, it's not nearly as good as normal CFOP or Roux. Also, future method ideas should probably be posted in the new method thread.

 

[brododragon]

Could an adapted version of Yau5 be used on 4x4 for Petrus? It would be the exact same except you don't solve the last cross edge and you make sure EO is correct while edge pairing. You could do OLL parity with a shortened alg because you don't have to preserve the Last Block, making parity a bit easier.

 

[Sub1Hour]

Could an adapted version of Yau5 be used on 4x4 for Petrus? It would be the exact same except you don't solve the last cross edge and you make sure EO is correct while edge pairing. You could do OLL parity with a shortened alg because you don't have to preserve the Last Block, making parity a bit easier.

I think that could work but I'm not sure if it's as good as existing methods for Petrus. I think that instead of doing a 3rd cross edge you could do the final edge in the beginning 2x2 and then move that over after so there is more freedom during a modified 323. I also think that doing something like this but with Hoya could also have some promise.

 

[ProStar]

Could an adapted version of Yau5 be used on 4x4 for Petrus? It would be the exact same except you don't solve the last cross edge and you make sure EO is correct while edge pairing. You could do OLL parity with a shortened alg because you don't have to preserve the Last Block, making parity a bit easier.

I've suggested this on here twice...

 

[brododragon]

I've suggested this on here twice...

I know but I wanted to add the forced EO skip and the easy parity part. Also, I wanted to give it some more publicity.

 

[brododragon]

I think that could work but I'm not sure if it's as good as existing methods for Petrus. I think that instead of doing a 3rd cross edge you could do the final edge in the beginning 2x2 and then move that over after so there is more freedom during a modified 323. I also think that doing something like this but with Hoya could also have some promise.

Can you elaborate? (Maybe a numbered list would help)

Here's something I came up with:

1. Hoya F4C (L/R centers, B/D centers).
2. Blockbuild a 1x3x4 on the D layer, and make sure it touches L/B/R, pairing dedged when needed.
3a. Pair 3 dedges simultaneously (like in 3-2-3), making sure to only pair one back slot.
3b. Do an F move, then pair up another 3 dedges. You should now have 3 dedges unpaired and 2x2x3 done.
3c. Somehow simultaneously do L3E and EO. Maybe an alg? I would imagine it wouldn't be too many.
OLL Parity: Use a shortened alg that doesn't preserve Last Block.
4. 3x3 stage with only 2-3 shortish steps! (LB, OLL, PLL or LB, ZBLL).


3x3 stage is massively faster than Yau 3x3 stage because F2L is only 2 pairs (plus an edge, but that takes no time at all), has super easy OLL, and then some normal PLL. The dedge pairing might still need some work; I'm no 4x4 expert.

EDIT: I forgot how to English.

 

[Sub1Hour]

Can you elaborate? (Maybe a numbered list would help)

Okay here we go

1. F2C // Standard F2C nothing fancy here
2. 2x2 edges // Make the 3 edges used in your 2x2 like you would in yau for cross in the same position as cross
3. L4C // Do this exactly like yau
4 Solve 2x2 // Solve the 2x2 block like you would in petrus
5 2-2-2-3 // Solve 2 edges using a 3 cycle and put those in the bottom slots, then proceed to do 3-2-3 but there is a pre-solved edge so its actually 2-2-3
6 3x3 stage // Self explanitory
6a Parity // Recognize OLL parity during your EO and use a shorter alg than the standard Drew or Lucas parity. PLL parity should be taken care of during PLL

If this method does not already exist I would like to coin the name of the method as Pau (Pronounced POW)

 

[brododragon]

Okay here we go

1. F2C // Standard F2C nothing fancy here
2. 2x2 edges // Make the 3 edges used in your 2x2 like you would in yau for cross in the same position as cross
3. L4C // Do this exactly like yau
4 Solve 2x2 // Solve the 2x2 block like you would in petrus
5 2-2-2-3 // Solve 2 edges using a 3 cycle and put those in the bottom slots, then proceed to do 3-2-3 but there is a pre-solved edge so its actually 2-2-3
6 3x3 stage // Self explanitory
6a Parity // Recognize OLL parity during your EO and use a shorter alg than the standard Drew or Lucas parity. PLL parity should be taken care of during PLL

If this method does not already exist I would like to coin the name of the method as Pau (Pronounced POW)

So in 3x3 you end up with XCross?

 

[Sub1Hour]

So in 3x3 you end up with XCross?

No. You have your 2x2 and then once you finish the reduction part then you make a 2x2x3, then do EO+Parity, Then you finish your F2L, OLL, And PLL like a regular Petrus solve.

 

[ProStar]

Okay here we go

1. F2C // Standard F2C nothing fancy here
2. 2x2 edges // Make the 3 edges used in your 2x2 like you would in yau for cross in the same position as cross
3. L4C // Do this exactly like yau
4 Solve 2x2 // Solve the 2x2 block like you would in petrus
5 2-2-2-3 // Solve 2 edges using a 3 cycle and put those in the bottom slots, then proceed to do 3-2-3 but there is a pre-solved edge so its actually 2-2-3
6 3x3 stage // Self explanitory
6a Parity // Recognize OLL parity during your EO and use a shorter alg than the standard Drew or Lucas parity. PLL parity should be taken care of during PLL

If this method does not already exist I would like to coin the name of the method as Pau (Pronounced POW)

I made something very similar to this, it was my first version of a Petrus/Petrus-W method for 4x4. If you go back, I'm sure you can find it

 

[xcross]

This idea probably exists already, and even if it doesn't, its horribly inefficent, but

1. first face -1 corner (face can be made of any 2 colors, but they must be opposite, ie: red and orange.)
2. CO
3. Seperate
4. PBL

 

[ProStar]

This idea probably exists already, and even if it doesn't, its horribly inefficent, but

1. first face -1 corner (face can be made of any 2 colors, but they must be opposite, ie: red and orange.)
2. CO
3. Seperate
4. PBL

This is Guimond

 

[FinnTheCuber]

Maybe new idea. Do a roux solve through eo. then transfer all of the edges to the correct face intuitively to get a normal pll case.

 

[ProStar]

Maybe new idea. Do a roux solve through eo. then transfer all of the edges to the correct face intuitively to get a normal pll case.

This has been proposed, F2B->OLL->B2E->PLL, it's not as good as plain CFOP or Roux

 

[FinnTheCuber]

Would anybody try it tho? I want someone whose better than me to see how fast they could get with it

 

[ProStar]

Would anybody try it tho? I want someone whose better than me to see how fast they could get with it

Trust me, people have tried it, and it's not as good. I got 23.49

 

[FinnTheCuber]

k

 

[Alexander]

On my new website i have made a view Blog post about some ideas that i have fopr a view puzzles like 3x3 and 4x4

3x3

• PLL headlight on the left
• 1SR (1 side recognition)

4x4

• building center and edges at the same time possible extetion for YAU

HOME | Speedcubing

Speedcubing ideas and methods this is the place for solving your rubiks cube.

apajoooms.wixsite.com

or

HOME | Speedcubing

Speedcubing ideas and methods this is the place for solving your rubiks cube.

sc.apajo.nl

 

[PapaSmurf]

Just in general on your website, it looks good but there are definitely a lot better PLL algs out there. Check these ones out. In terms of PLL recognition, that's already a commonly used thing but your OLL 1SR would definitely be useful for someone trying to seriously optimise their LL. The 4x4 yau thing isn't very clear, but from the dates on the website I can tell that your native language isn't English and is probably Dutch. From what I can gather though, it's just influencing the edges so that when you solve the centres you skip the edge pairing step, which would definitely be a useful trick in some cases.