The New Method / Substep / Concept Idea Thread

[Deleted member 36445]

I came up with a last slot method for 3x3. I call it GLS

1.F2l-1
2. Insert corner.
3. insert the edge while orienting all LL edges.
4.LL

I will start genning algs for this.
If this has already been invented, just tell me.

there is MGLS by @Lucas Garron where you insert edge orienting edges, inser corner orienting corners, and do PLL

 

[Thom S.]

1.F2l-1

That is one big step

If this has already been invented, just tell me.

It has

 

[G DA CUBER]

here is MGLS by @Lucas Garron where you insert edge orienting edges, inser corner orienting corners, and do PLL

In this substep, you do the corner first then you do the edge.

Also, I am aware of MGLS

 

[Deleted member 36445]

In this substep, you do the corner first then you do the edge.

and what's the advantage?

also MGLS ends in PLL, 21 algs, and your method ends with ZBLL which is a lot more

 

[G DA CUBER]

The 'please read before posting' at the start does not specify that the method has to be good.

 

[Thom S.]

The 'please read before posting' at the start does not specify that the method has to be good.

1. Get out.
2. It does, however ask you to think about your method for more that 6 Minutes, which you clearly didn't do.

 

[Deleted member 36445]

The 'please read before posting' at the start does not specify that the method has to be good.

bruh, no one specified to me that I have to breath to stay alive. I'm just not plain dumb

 

[DuckubingCuber347]

The 'please read before posting' at the start does not specify that the method has to be good.

So then why are you proposing a method that's not good or original? APB is a great method proposal, RUBAR is an original method. Yours is a method anyone can think up. This thread already has enough method dumps. If you're going to chide people for criticizing your method at least make the method good.

 

[PapaSmurf]

I came up with a last slot method for 3x3. I call it GLS

1.F2l-1
2. Insert corner.
3. insert the edge while orienting all LL edges.
4.LL

I will start genning algs for this.
If this has already been invented, just tell me.

It is just ZB with more steps.

 

[levi.m]

random Rubik's cube method
1.Solve a roux block + DB edge
2.solve the down front edge and solve the downright edge at the same time
3.solve the back F2L pair while pairing your last F2L pair
4.BLS (Bingus last slot) basically VLS or HLS but you solve the corners while solving oll
5.PLL

 

[Thom S.]

random Rubik's cube method

I'm telling you guys, if the first line of a post reads like this, the method was made in 5 minutes.

random Rubik's cube method
1.Solve a roux block + DB edge
2.solve the down front edge and solve the downright edge at the same time
3.solve the back F2L pair while pairing your last F2L pair
4.BLS (Bingus last slot) basically VLS or HLS but you solve the corners while solving oll
5.PLL

You didn't solve the M Centers.
BLS should exist somewhere. The algs are better that ELSOLL, but are longer than VLS.

 

[levi.m]

I'm telling you guys, if the first line of a post reads like this, the method was made in 5 minutes.

You didn't solve the M Centers.
BLS should exist somewhere. The algs are better that ELSOLL, but are longer than VLS.

i made it in five days

 

[Deleted member 36445]

i made it in five days

1 min a day

 

[DuckubingCuber347]

random Rubik's cube method
1.Solve a roux block + DB edge
2.solve the down front edge and solve the downright edge at the same time
3.solve the back F2L pair while pairing your last F2L pair
4.BLS (Bingus last slot) basically VLS or HLS but you solve the corners while solving oll
5.PLL

Another way you can say it is:
1. xxcross
2. Do BdR really bad so LS can be good
3. nvm make LS bad so you can get an OLL skip (Literally VLS)
4. PLL

Not the worst method in the world but it's still not great. At all.

 

[sDLfj]

I thought of a fast way to solve the last layer. I'm sure it's already been thought of, but I'm just wondering how fast it is compared to CFOP 3lll (which I currently know). Solve the corners (orient and permutate) with 2x2 cll. Then use <M,U> algorithms to solve the cube.

These averages assume all cases have an equal probability of appearing
The average moves for CLL(10.02) + <M,U> Perms (7.5) is 17.5 moves
The average moves for PLL are 13.4

OLL is likely slower than CLL, and <M,U> perms are faster than the average PLL. So solving the corners, then edges, should be faster than solving OLL, then PLL.

 

[NigelTheCuber]

I thought of a fast way to solve the last layer. I'm sure it's already been thought of, but I'm just wondering how fast it is compared to CFOP 3lll (which I currently know). Solve the corners (orient and permutate) with 2x2 cll. Then use <M,U> algorithms to solve the cube.

These averages assume all cases have an equal probability of appearing
The average moves for CLL(10.02) + <M,U> Perms (7.5) is 17.5 moves
The average moves for PLL are 13.4

OLL is likely slower than PLL, and <M,U> perms are faster than the average PLL. So solving the corners, then edges, should be faster than solving OLL, then PLL.

that's cll+ell and many people know what it is. it is slower than normal oll+pll

 

[sDLfj]

that's cll+ell and many people know what it is. it is slower than normal oll+pll

Okay, is it slower because of cll? I would have thought oll and cll would be similar (and cll would be quicker) and with how much faster ell is compared to pll, the cll+ell would be faster than oll+pll. Do you think for 3lll, cll+ell is better?

Nevermind: I only just realized that this method wouldn't work as well as I thought. I (for some reason) thought that the edges could be solved with the <M,U> perms, when they would still also need oriented correctly. The algs to do this (ell) aren't as fast as the <M,U> perms.

 

[Deleted member 36445]

Okay, is it slower because of cll? I would have thought oll and cll would be similar (and cll would be quicker) and with how much faster ell is compared to pll, the cll+ell would be faster than oll+pll. Do you think for 3lll, cll+ell is better?

Nevermind: I only just realized that this method wouldn't work as well as I thought. I (for some reason) thought that the edges could be solved with the <M,U> perms, when they would still also need oriented correctly. The algs to do this (ell) aren't as fast as the <M,U> perms.

CLL

CxLL - Speedsolving.com Wiki

ELL

https://sarah.cubing.net/3x3x3/ell.pdf

 

[DuckubingCuber347]

Okay, is it slower because of cll? I would have thought oll and cll would be similar (and cll would be quicker) and with how much faster ell is compared to pll, the cll+ell would be faster than oll+pll. Do you think for 3lll, cll+ell is better?

Nevermind: I only just realized that this method wouldn't work as well as I thought. I (for some reason) thought that the edges could be solved with the <M,U> perms, when they would still also need oriented correctly. The algs to do this (ell) aren't as fast as the <M,U> perms.

For a lot of ELL cases <MU> isn't optimal, <S,R,U,M> would be better (Though obviously, you won't have all of that in very many algs.)

The method you proposed is called CFCE. The reason it isn't as popular is due to many reasons but it is often considered inferior because recognition is worse and the algorithms are trickier. It's certainly not an invalid method, it's just few cubers want to take the risk.

ELL

https://sarah.cubing.net/3x3x3/ell.pdf

Don't even bother.

ELL algorithms

These are much better.

 

[lucazdoescubingstuff]

so i thought of a 3x3 method that i think would be very useful

1. You first make 2 1x2x3 blocks like in Roux
2. Solve CMLL
3. insert the remaining edge of the face that you made the 1x2x3 blocks with
4. you get less OLLs if you know full OLL