The New Method / Substep / Concept Idea Thread

[Thumbprint]

Back with another probably bad method idea:
Step 1. Solve a face plus 2 f2l pairs on the left
Step 2. Do coll to solve the top corners
Step 3. Do an algorithm to solve the UL edge (or more; I just came up with this and don't know how to gen algs)
Step 4. Do another algorithm to solve the rest of the cube (still don't know how to gen algs so this also might not be possible)

 

[abunickabhi]

Back with another probably bad method idea:
Step 1. Solve a face plus 2 f2l pairs on the left
Step 2. Do coll to solve the top corners
Step 3. Do an algorithm to solve the UL edge (or more; I just came up with this and don't know how to gen algs)
Step 4. Do another algorithm to solve the rest of the cube (still don't know how to gen algs so this also might not be possible)

This sounds super similar to Roux, the FB and LSE step are a bit different.

 

[Thumbprint]

This sounds super similar to Roux, the FB and LSE step are a bit different.

Thinking about it now, it does sound similar to roux. I came up with this after learning commutators and I thought that if I made a layer plus 2 f2l pairs I could solve the rest of the left side using commutators, but then it hit me that I could just do coll and possibly come up with an alg set that would solve the UL edge, then another alg set to solve the last 5 edges. But it is very similar to roux.

 

[Megaminx lover]

random idea
1. F2B
2. EODFDB
3. ZBLL
Basically a roux based method that is marginally better than roux for at least 400 more algs

 

[RyanSoh]

random idea
1. F2B
2. EODFDB
3. ZBLL
Basically a roux based method that is marginally better than roux for at least 400 more algs

https://www.speedsolving.com/wiki/index.php/ZBRoux

 

[Megaminx lover]

https://www.speedsolving.com/wiki/index.php/ZBRoux

oof.

 

[Cubing Forever]

random idea
1. F2B
2. EODFDB
3. ZBLL
Basically a roux based method that is marginally better than roux for at least 400 more algs

What @RyanSoh said and it's worse than standard Roux(@Tao Yu is more qualified to give an opinion on this lol)

 

[LukasCubes]

I proposed this method on my 1LLS method doc and in discord but ima finally do it here now

Triangular Waterman

Step 1: Triangle like in triangular Francisco on BDL (Average Movecount: ~7)
Step 2: 1LLS, 615 cases I found out recently (Average Movecount: ~8)
Step 3: Solve another 3-6 Edges, relies on intuition and TPS may not be high here. (Average Movecount: ~11)
Step 4: Waterman L4E, L5E, L6E, or L7E, after performing algs, you permute midges. (Average Movecount: ~14)

Step 1 can be used for blockbuilding intuition. Step 2 has 615 algs for 1LLS with the subgroups being CLL, TCLL+, TCLL-, TTLL, A bunch of RU gen algs, ones where the last 1st layer corner is oriented, and the rest. Step 3 can be used for intuition again, but can use algs to solve 2-3 edges at one alg. 1LL4E and 1LL5E combined is like 270-280 algs and L6E and L7E like normal (not 1 look) is almost 400 algs and the 3 move 4 algs for permuting midges.

7+8+11+14=40

~40 AVERAGE MOVECOUNT LOL

between 1000-1200 algs tho thats bad.

 

[Filipe Teixeira]

f2l-1
epls (els + epll)
l5c

 

[Thom S.]

f2l-1
epls (els + epll)
l5c

I'm extremely sure this exists and Algorithms are generated

 

[Cubing Forever]

I'm extremely sure this exists and Algorithms are generated

it's 8355 but advanced

 

[WhyDoINeedAUsername]

This is my first post on here so sorry if I get anything wrong.

I didn't put new in the title or anything for this method as I'm unsure as to if it has already been developed, I'm aware of the speedcubing wiki and all the pages of methods It has so I have checked as hard as I can to see if I can find this method.

The methods name is snow, it used to stand for something but now it doesn't mainly because I forgot what it stands for however I'm still going to use it if it is a new discovery.

A few things to say about this method is it should have a minimum of 4 algorithms more to learn if you know CFOP and is also very similar to CFOP except an extra step in between F2L and OLL. My current Average PB is on this method but my single PB is using CFOP, average used to be on CFOP and single used to be with this method until about 4 days ago

Steps:
Cross Minus an edge
F2L Minus a pair, The unsolved cross piece allows for a lot more trick with this step so take advantage of it (There are only 2 pairs in every solve that you don't have to insert, the other two you do have to insert)
Create and insert the final 1x2x2 block (i'll explain this one soon and leave some algorithms that can be used)
OLL (Other algorithm sets can be used as well)
PLL (Other algorithm sets can be used as well)

For F2L minus a pair, the only pairs that you don't solve is one of the two that go next to the cross piece edge that isn't inserted, so for example I'm using white cross and I didn't insert the blue edge. This means I either need to insert the blue and orange pair or the blue and red pair, Which ever pair you insert also decides on which algorithm you do afterwards but they are all very similar. If you chose to not use the blue white edge in the cross you have to insert the green red pair and green orange pair

This next step is one I hope people can figure out as I don't think I will right it well, you want to make you're last pair but before inserting it you wanna insert the missing cross piece into the place you would usually put the last pair, to check if its the correct orientation do r U r' or l' U' l depending on the pair you inserted to see if the edge will solve the cross, if not you want to flip that edge (you dont actually insert and uninsert the edge again in the method, this is just for the first time so people figure out what orientation you are meant to put the edge in, this can be figured out easily during normal solves afterwards. While inserting the edge make sure you don't break up the pair, I hope this can be figured out without me needing to explain it. Once you have done that bring the F2L piece the front most likely with a U2, there will now be 4 algorithms, fixed box a, fixed box b, broken box a and broken box b.

Fixed Box A ( if the Block is made on the left side of the cube)

l' U' l

Fixed Box B (Block on the right side)

r U r'

Broken Box A (Block needs to be made on the left, but pair is on right side instead)

U' F U F' L' U2 L U' F' L F L' U2 l' U' l
L F' L' F2 U F2 L F L' U2 l' U' l

Broken Box B (Block needs to be made on the right side)

U F' U' F R U2 R' U F R' F' R U2 r U r
R' F R F2 U' F2 R' F' R U2 r U r'

For broken box, the first algorithm is easier to me and I believe the second is faster. After this you have OLL and PLL or CLL and ELL or any thing else you want to learn.

So is this method new, been developed previously or too similar to another method to be considered its own method. If its new I would recommend learning it and giving a go, it would be nice to see what good cubers get as my PB is 18 seconds, I wouldn't recommend changing from any of your main methods unless you truly prefer this one though.

I would just like to point out that my old PB was using an older version of this method which was making full F2L and then just inserting the edge in afterwards, im actually unsure as to which version of the method is better

I've just done a little testing and I would say use the block part if you are fast enough with it, if not solve all F2L and insert the edge in after. One more thing you can do is solve the edge during F2L which can happen sometimes by accident or solve the edge and then insert the final F2L piece.

 

[EvanCuber]

Yo I'm back with two more most likely stupid ideas, this time for 3BLD.

Belt Method
Okay, I know it sounds dumb. But for people who don't want to learn a ton of algorithms yet still be decent at three blind, why not just do the belt, and then, ES (Edge Separation), CS (Corner Separation), FP(Fix All Parities), OLL, PLL. All you have to do is one look your belt and remember how each algorithm moves your pieces.

Anti-Belt Method
Probably sounds even dumber but what if you separated all your edges to their correct side while except the E slice. Then you fix all Parities, Permute U and D layer edges with 144 algorithms or 2 steps of 12 algorithms, then separate corners, Permute U layer corners, Permute D layer corners. Last, L4E, to solve the E slice. On this method you have to one look ESEO, and remember how each algorithm affects each piece.

 

[cuberswoop]

Quick overview:

COELL (so-ell)

3C - Make 3 out of 4 of the cross edges (3Cross)

RMO - All of F2l Done except for 1 cross piece (Roux Minus One)

CPE - Orient all the last edges while inserting the last cross piece. (Cross Plus Edges)

OLL - Orient the last layer (You can use ZBLL, COLL, CLL, and lots of things here since all the edges are up)

PLL - Permute The Last Layer


This is an improved version of CFOP, it uses mostly R, U, M, and L moves. It's very ergonomic, and lookahead is the same, if not easier, then cfop. The algs were made with cube explorer. I have a W.I.P google doc (https://docs.google.com/document/d/1Q3fV-Br-41YNF42mA2IakGuniDkazBdm4EK2lQb5deQ/edit) with some algorithms to orient the last edges and insert the last cross edge.

Spoiler: Solve

Scramble: U2 B2 F' D2 U2 L2 U2 B R2 F' R2 B2 U' B2 U' F' D2 B L2 R' F2'

Solve:
Inspection: z2

3C: R' L' D2

RMO1: R U R' D R U' R' D'
RMO2: U' R U2 R' U' R U2 R'
RMO3: y2 U' L' U L U F U' F'
RMO4: U R U R' U2 R U R'

CPE: y2 F2 D2 F' M2 F E2 B' U M2 U B2
OLL: Skip
PLL: M2 U M U2 M' U M2 U2

Please give feedback on my method and tell me if it's already been invented.

 

[ruffleduck]

This is just a worse version of HK.

 

[DuckubingCuber347]

Quick overview:

COELL (so-ell)

3C - Make 3 out of 4 of the cross edges (3Cross)

RMO - All of F2l Done except for 1 cross piece (Roux Minus One)

CPE - Orient all the last edges while inserting the last cross piece. (Cross Plus Edges)

OLL - Orient the last layer (You can use ZBLL, COLL, CLL, and lots of things here since all the edges are up)

PLL - Permute The Last Layer


This is an improved version of CFOP, it uses mostly R, U, M, and L moves. It's very ergonomic, and lookahead is the same, if not easier, then cfop. The algs were made with cube explorer. I have a W.I.P google doc (https://docs.google.com/document/d/1Q3fV-Br-41YNF42mA2IakGuniDkazBdm4EK2lQb5deQ/edit) with some algorithms to orient the last edges and insert the last cross edge.

Spoiler: Solve

Scramble: U2 B2 F' D2 U2 L2 U2 B R2 F' R2 B2 U' B2 U' F' D2 B L2 R' F2'

Solve:
Inspection: z2

3C: R' L' D2

RMO1: R U R' D R U' R' D'
RMO2: U' R U2 R' U' R U2 R'
RMO3: y2 U' L' U L U F U' F'
RMO4: U R U R' U2 R U R'

CPE: y2 F2 D2 F' M2 F E2 B' U M2 U B2
OLL: Skip
PLL: M2 U M U2 M' U M2 U2

Please give feedback on my method and tell me if it's already been invented.

I don't really get the example solve. y2 rotations are horrible and two of them are really awful, the CPE had pretty bad moves like B and E moves, and cross really isn't that hard. If you don't want to do cross just use Nautilus/Roux/Mehta and if you really want to do something like this just use HK (https://www.speedsolving.com/wiki/index.php/Hawaiian_Kociemba)

 

[LukasCubes]

This is just a worse version of HK.

better than CFOP isnt it?

Quick overview:

COELL (so-ell)

3C - Make 3 out of 4 of the cross edges (3Cross)

RMO - All of F2l Done except for 1 cross piece (Roux Minus One)

CPE - Orient all the last edges while inserting the last cross piece. (Cross Plus Edges)

OLL - Orient the last layer (You can use ZBLL, COLL, CLL, and lots of things here since all the edges are up)

PLL - Permute The Last Layer


This is an improved version of CFOP, it uses mostly R, U, M, and L moves. It's very ergonomic, and lookahead is the same, if not easier, then cfop. The algs were made with cube explorer. I have a W.I.P google doc (https://docs.google.com/document/d/1Q3fV-Br-41YNF42mA2IakGuniDkazBdm4EK2lQb5deQ/edit) with some algorithms to orient the last edges and insert the last cross edge.

Spoiler: Solve

Scramble: U2 B2 F' D2 U2 L2 U2 B R2 F' R2 B2 U' B2 U' F' D2 B L2 R' F2'

Solve:
Inspection: z2

3C: R' L' D2

RMO1: R U R' D R U' R' D'
RMO2: U' R U2 R' U' R U2 R'
RMO3: y2 U' L' U L U F U' F'
RMO4: U R U R' U2 R U R'

CPE: y2 F2 D2 F' M2 F E2 B' U M2 U B2
OLL: Skip
PLL: M2 U M U2 M' U M2 U2

Please give feedback on my method and tell me if it's already been invented.

use L5EOP instead of just pick better algs for CPE

 

[EvanCuber]

CPE - Orient all the last edges while inserting the last cross piece. (Cross Plus Edges)

OLL - Orient the last layer (You can use ZBLL, COLL, CLL, and lots of things here since all the edges are up)

Or you can just use ZZ, which is the best method for EO. Also the algorithm you used for CPE in your example solve would literally take the fastest speed solvers half their solve; it is just ergonomically terrible. Too many methods have been created along the line of leaving out a cross edge, another being COLL and then L5E.

 

[LBr]

better than CFOP isnt it?

better than waterman xD

 

[EvanCuber]

better than waterman xD

Let's not start that in this thread too. To think it was already bad enough in the others