The New Method / Substep / Concept Idea Thread

[efattah]

We've already established that several methods can get sub-40 move averages, if you are willing to learn lots of algs. I posted how LMCF could get sub-40 average moves with 4000 algorithms.

If someone can find an effective speedsolving method that gets sub-40 moves in less than 500 algorithms that would be revolutionary.

 

[PetraPine]

New method idea for 2x2

Step1:Solve the of the first layer corners,

Step2: Orient any other corner into the fourth spot (if it is the first layer corner just do a cll to finish the solve)

Step3 Do a pll to orient top layer and put the white corner in the back and the incorrect corner of the first layer in the front and do one of six TTLL
Algorithms(corners only)

this method seems like a good intermediate method, and if you learn the full set of tols(orients all 5 last corners even if front corner is flipped)
than it could be pretty efficient.

Ribbon Method Printable Algorithm Sheets

docs.google.com

this is a list of the ttlls/tols from the ribbon method.

EXAMPLE SOLVE:
scramble:R F R' F' U2 F R'
Z' R' U2 R Y' L' U' L U L F' L' F U2 (alg) D U' R2 U R2 U' R2 U' R2 U

 

[PapaSmurf]

We've already established that several methods can get sub-40 move averages, if you are willing to learn lots of algs. I posted how LMCF could get sub-40 average moves with 4000 algorithms.

If someone can find an effective speedsolving method that gets sub-40 moves in less than 500 algorithms that would be revolutionary.

This.

Instead of working separately and sparadically we could possibly do something that's a pseudo-combined effort. I, for one, am a fan of FB then 2x2x2 (SSF2L). Working together could mean more progress quickly.

 

[Cubingcubecuber]

New method idea for 2x2

Step1:Solve the of the first layer corners,

Step2: Orient any other corner into the fourth spot (if it is the first layer corner just do a cll to finish the solve)

Step3 Do a pll to orient top layer and put the white corner in the back and the incorrect corner of the first layer in the front and do one of six TTLL
Algorithms(corners only)

this method seems like a good intermediate method, and if you learn the full set of tols(orients all 5 last corners even if front corner is flipped)
than it could be pretty efficient.

Ribbon Method Printable Algorithm Sheets

docs.google.com

this is a list of the ttlls/tols from the ribbon method.

EXAMPLE SOLVE:
scramble:R F R' F' U2 F R'
Z' R' U2 R Y' L' U' L U L F' L' F U2 (alg) D U' R2 U R2 U' R2 U' R2 U

This is basically VOP

 

[RedstoneTim]

New method idea for 2x2

Step1:Solve the of the first layer corners,

Step2: Orient any other corner into the fourth spot (if it is the first layer corner just do a cll to finish the solve)

Step3 Do a pll to orient top layer and put the white corner in the back and the incorrect corner of the first layer in the front and do one of six TTLL
Algorithms(corners only)

this method seems like a good intermediate method, and if you learn the full set of tols(orients all 5 last corners even if front corner is flipped)
than it could be pretty efficient.

Ribbon Method Printable Algorithm Sheets

docs.google.com

this is a list of the ttlls/tols from the ribbon method.

EXAMPLE SOLVE:
scramble:R F R' F' U2 F R'
Z' R' U2 R Y' L' U' L U L F' L' F U2 (alg) D U' R2 U R2 U' R2 U' R2 U

The approach you're outlining here is similar to the HD method, except that corners are oriented less efficiently, which in turn doesn't allow you to do PBL (unless one uses your advanced version with TOLS, which would be exactly the same as standard HD).
A small tip I would give you is not to use 3x3 algorithms for 2x2 since, as you may have noticed, they're a bit long and move sequences like U D' don't make much sense on 2x2.

 

[brododragon]

This.

Instead of working separately and sparadically we could possibly do something that's a pseudo-combined effort. I, for one, am a fan of FB then 2x2x2 (SSF2L). Working together could mean more progress quickly.

I think that M-CELL could do great as a base to try to develop a method like this. It's got SSF2L and a very efficient, yet low alg finish. I think with non-matching blocks you could probably get under 40. It might sound extremely hard to use NMB, but I think it would be something like blind, where it's mind breaking the first few times, but eventually becomes easier. Some other things you could do: 1-look 2x2x3 (or at least two-look), L6C, L7E.

 

[FinnTheCuber]

New idea. Create your last f2l pair but don't insert it. Then you can use it to intuitively force an edges oriented or case for I'll or a skip. If this is a new idea I will go into more detail in a separate post.

 

[ProStar]

New idea. Create your last f2l pair but don't insert it. Then you can use it to intuitively force an edges oriented or case for I'll or a skip. If this is a new idea I will go into more detail in a separate post.

This is just VHLS except done intuitively, which is just ZBLS except the pair has to be made

 

[FinnTheCuber]

Is it a new execution tho?

 

[ProStar]

Is it a new execution tho?

No it's just doing it intuitively, which I have already done(I don't do it in solves because most cases aren't that good)

 

[FinnTheCuber]

Ok

 

[PetraPine]

The approach you're outlining here is similar to the HD method, except that corners are oriented less efficiently, which in turn doesn't allow you to do PBL (unless one uses your advanced version with TOLS, which would be exactly the same as standard HD).
A small tip I would give you is not to use 3x3 algorithms for 2x2 since, as you may have noticed, they're a bit long and move sequences like U D' don't make much sense on 2x2.

I thought there werent any algs like this for 2x2 I was just sumplaminting these for now
Also it isn't exactly because you're orienting all the corners first and than doing a ttll.
I should of done my research before posting this as I didn't know of the hd method

 

[Sub1Hour]

I'm not sure if this alg set exists yet but I'm gonna throw the idea out there just in case it doesn't. COBL. It's a square-1 set that combines EO and CP into one alg. Not sure how many algs there will be but I think this could potentially be a half-decent way to do 4 look solves without learning PBL.

 

[PapaSmurf]

It apparently exists but the algs are apparently bad. That's from a conversation I had with Charlie Stark. Also, 2 alg PBL exists now so you're better off doing that.

 

[RedstoneTim]

I thought there werent any algs like this for 2x2 I was just sumplaminting these for now
Also it isn't exactly because you're orienting all the corners first and than doing a ttll.
I should of done my research before posting this as I didn't know of the hd method

In case you want some NLL algorithms, here are the "official" ones, WoowyBaby's algs and a sheet from someone else whose name I won't even try to spell.
I wouldn't recommend also permuting corners for your V since everything becomes a lot less efficient and full NLL itself isn't that many algorithms.

 

[KingOfThePanda]

Is This A Method Used By Anyone?

Let me preface this with letting you know that I literally started attempting to speed solve last week, so if this is stupid, just let me know and let me know why.

Currently I am using CFOP, and have a very small OLL and PLL repertoire. I'm averaging 49 seconds out of sheer willpower alone. The biggest issue (besides optimal F2L building) is getting an efficient cross solution.

And because I suck at the Cross in CFOP, I recently started just putting White on bottom and doing F2L, completely ignoring the white cross pieces. After that, I solve the White edges. I asked the Discord, and someone mentioned that this was pretty much just a method called "FreeFOP" (I may have got that name wrong)

Well, I've been messing around with it and I've found that when you just ignore the Cross and do those F2L pairs, the yellow corner pieces on top can be solved just like normal. Then you just have to only use M and U moves to solve the bottom white layer. So now you have F2L solved and the yellow corners oriented. The next step is to figure out a way to move around the yellow edges with only M and U moves.

Firstly, is this an actual method, or am I just bad for ignoring my cross problems?

If this is a method, then how do I find information on how to get better at it? I feel way more comfortable doing it this way, but if there is a good reason why this is bad, then I have no choice but to move on and crunch some crosses.

Any help or advice would be appreciated!

Oh, also I was thinking about it and if you basically just solve only 2 opposite cross pieces in the beginning and do this, you have a Roux setup. So maybe I should learn Roux.

 

[shadowslice e]

It sounds like you're halfway to developing a bad version of something like ssc or pcms.

So, yes it is a known method (or at least a method which has been thought about and considered in the past) but you are still bad for ignoring your cross pieces since that adds so many additional moves and ruins the ergonomics of your solves.

 

[maticuber]

It sounds like a bad version or corner first (basically you solve corners like in a 2x2, then edges). That method was dominant in the 80s, but modern versions of cfop are better.

So, the biggest issue with that method are blindspots, Cross+F2L is great because you already know the location of 1/4th of your edges, and after inserting your first F2L pair you know the location of all the pieces. Not doing the cross takes away a lot of information, not only BD is out of sight, but also all the D face stickers.

You have to also take into consideration that when you do cross, you use 8 moves or less to solve 4 pieces, and after F2L you have 8 pieces that needs to be solved and OLL+PLL does it in ~21 moves. With your method you'll have 12 pieces that needs to be solved and you can't see.

An alternative method that's really similar to what you are trying to do is roux. You completely ignore the cross but do block building, imagine F2L without the M slice, after that you solve the corners and the cube can be solved using M U moves (I'm not familiar with Roux, maybe someone else can explain the method).

 

[AlphaCuber is awesome]

Let me preface this with letting you know that I literally started attempting to speed solve last week, so if this is stupid, just let me know and let me know why.

Currently I am using CFOP, and have a very small OLL and PLL repertoire. I'm averaging 49 seconds out of sheer willpower alone. The biggest issue (besides optimal F2L building) is getting an efficient cross solution.

And because I suck at the Cross in CFOP, I recently started just putting White on bottom and doing F2L, completely ignoring the white cross pieces. After that, I solve the White edges. I asked the Discord, and someone mentioned that this was pretty much just a method called "FreeFOP" (I may have got that name wrong)

Well, I've been messing around with it and I've found that when you just ignore the Cross and do those F2L pairs, the yellow corner pieces on top can be solved just like normal. Then you just have to only use M and U moves to solve the bottom white layer. So now you have F2L solved and the yellow corners oriented. The next step is to figure out a way to move around the yellow edges with only M and U moves.

Firstly, is this an actual method, or am I just bad for ignoring my cross problems?

If this is a method, then how do I find information on how to get better at it? I feel way more comfortable doing it this way, but if there is a good reason why this is bad, then I have no choice but to move on and crunch some crosses.

Any help or advice would be appreciated!

Oh, also I was thinking about it and if you basically just solve only 2 opposite cross pieces in the beginning and do this, you have a Roux setup. So maybe I should learn Roux.

this sounds very similar to PCMS which while isn’t terrible but also isn’t as good as roux. if you want a good method that gives you a very free f2l I would advise roux. my other suggestion is to look into lots of different corners first methods such as PCMS and try and develop a viable method for modern speedsolving. If you try this it will take a lot of work and you will spend lots of time on something that may go knowhere But you will get to develop new things and may find the next big method.

 

[xcross]

this is 2 cross pieces away from being roux essentialy.