The New Method / Substep / Concept Idea Thread

[Teoidus]

Ah alright, I suppose i can try genning algs as well then. Is there any reason it's being kept under wraps?

 

[Shiv3r]

hey efattah so for LCMF what are the edge permutation algorithms or the other algorithms? I mean for a corners first method this is pretty fast, Im hyped.

 

[gyroninja]

Ah alright, I suppose i can try genning algs as well then. Is there any reason it's being kept under wraps?

He wants the method to be really polished when it's released. Generating tsle is pretty easy. I would recommend using the qtm version of cube explorer. You notice how good the algs can be while being almost optimal move wise.

 

[wir3sandfir3s]

Kind of random idea. Guess it can be called "Columns Last". Probably won't be very efficient, but whatever.
1. EO?
2. Solve M slice.
3. Solve S slice.
4. Use an alg to solve FL corners.
5. Finish the L4P (last 4 pairs).
Again, just a random thought I had.

 

[Teoidus]

He wants the method to be really polished when it's released. Generating tsle is pretty easy. I would recommend using the qtm version of cube explorer. You notice how good the algs can be while being almost optimal move wise.

How do you think this method compares to mainstream ones? (CFOP, Roux, ZZ)

 

[4Chan]

How do you think this method compares to mainstream ones? (CFOP, Roux, ZZ)

It's basically a gambler's method.

At the highest level of cubing, it can give absolutely ridiculous singles due to the 20% OLL skip, 1/360 "true" LL skip, and absurdly short/fast algs like triple sledge. The combination of a short TSLE and TTLL would approximate an non-forced LL skip caused by a bad F2L pair.

But in terms of averages, the TSLE step takes as much time as OLL with ZZ, and the TTLL is roughly equivalent to PLL.
At the very most, it only saves as much time as it takes to do that slot.

So at the highest level, just imagine ZZ with one less slot, and some sick singles.
(Which, I guess might be better than normie ZZ, but with normie ZZ, there's tons of tricks like ez ZBLL, CLS, influence tricks, and stuff that might make it a moot point.)

Honestly, if the pair is pre-made or R U R', I'm inserting it and doing ZBLL.

 

[wir3sandfir3s]

It's basically a gambler's method.

At the highest level of cubing, it can give absolutely ridiculous singles due to the 20% OLL skip, 1/360 "true" LL skip, and absurdly short algs like triple sledge. The combination of a short TSLE and TTLL would approximate an non-forced LL skip caused by a bad F2L pair.

But in terms of averages, the TSLE step takes as much time as OLL with ZZ, and the TTLL is roughly equivalent to PLL.
At the very most, it only saves as much time as it takes to do that slot.

So at the highest level, just imagine ZZ with one less slot, and some sick singles.
(Which, I guess might be better than normie ZZ, but with normie ZZ, there's tons of tricks like ez ZBLL, CLS, influence tricks, and stuff that might make it a moot point.)

Honestly, if the pair is pre-made or R U R', I'm inserting it and doing ZBLL.

Is it worth switching to from FreeFOP?

 

[2180161]

I like that idea, and I think we can add something to help influence more so EJLS won't be all that is influencing the Orientation.

So what if we changed it to:

I. 2x2x2 in DBL + Slight Influence on EO (For 6+ Bad edges)
II. 2x2x3 + EO + Set up easy Pseudo CE Pair
III. Insert the Pseudo CE Pair into an F2L-Slot while Orienting all of the Corners + Placing E-Slice Edges (One will already be in place from the CE Pair)
IV. CP+EP

This would have more algorithms, but I think it would be more efficient, influencing throughout most of the solve.

Example Solve:
x' z //Inspection
D2 l U2 u' L //2x2x2 + Orientation Influence
R2 U l' D R' D x' //2x2x3 + EO + E-Slice + Some Oriented Corners (Lucky)
R' U2 R U2 R2 U' R' U' R U R2 //CO + 1x1x2 (Cancelled 'R' Move)
D R' U' R D' R U R D' R U' R' D R //CP+EP
36 STM

Used the same scramble as Crafto.
I did something slightly different because of the D-Layer Orientation Skip.
I think it can have decent fingertricks, after 2x2x3 at least, 2G+3G Steps.

This is a very similar finish to HTA if I understand correctly. Do I understand correctly?
Also the amount of algs would be huge. If I understand correctly, my math is bad though, it would be about alot of algs.

 

[JTWong71]

This is a very similar finish to HTA if I understand correctly. Do I understand correctly?
Also the amount of algs would be huge. If I understand correctly, my math is bad though, it would be about alot of algs.

This is a similar version to ECE (Or SSC, or AB, all the same methods), which is a "Sub-Optimal" Method to HTA.
I don't know how many algorithms there would be, as there are 3 more pieces to solve along with no AUF's for symmetric cases due to Non-U Layer Pieces.

 

[Shiv3r]

Hey Guys, I came up with an Idea for doing a 1-look corners(or something that requires both orientation and permutation algorithms) that is very alg-efficient, and could be also taught to noobs as a 2-look. first is to have an OCLL algset that doesn't affect the positions of the corners(like a 2-gen OCLL that restores the positions after) and the regular permutation algorithms. that means that you can 1-look both the orientation and permutation case before you solve them, you just remember, the OLL and the PLL case(and any AUF's in between). This would be useful for methods like corners-first, ortega, roux, pikashit, etc.
recognition would be the same as it is for 1-look corners, i think.

so, what do you think? executing 2 algorithms in 1 look? a good idea?

 

[shadowslice e]

Hey Guys, I came up with an Idea for doing a 1-look corners(or something that requires both orientation and permutation algorithms) that is very alg-efficient, and could be also taught to noobs as a 2-look. first is to have an OCLL algset that doesn't affect the positions of the corners(like a 2-gen OCLL that restores the positions after) and the regular permutation algorithms. that means that you can 1-look both the orientation and permutation case before you solve them, you just remember, the OLL and the PLL case(and any AUF's in between). This would be useful for methods like corners-first, ortega, roux, pikashit, etc.
recognition would be the same as it is for 1-look corners, i think.

so, what do you think? executing 2 algorithms in 1 look? a good idea?

Why not just learn COLL? The recognition is probably better, the algs are shorter and can be executed faster.

This is a similar version to ECE (Or SSC, or AB, all the same methods), which is a "Sub-Optimal" Method to HTA.
I don't know how many algorithms there would be, as there are 3 more pieces to solve along with no AUF's for symmetric cases due to Non-U Layer Pieces.

ABC isn't sub-optimal HTA. In fact, I'm pretty sure that it is more efficient than HTA. Do you mean just the regular Thistlethwaite/Kociemba Algorithm?

 

[Shiv3r]

I figured it out: im going to use the CMLL orientation-only algorithms. that should help.

 

[JTWong71]

ABC isn't sub-optimal HTA. In fact, I'm pretty sure that it is more efficient than HTA. Do you mean just the regular Thistlethwaite/Kociemba Algorithm?

Yeah, I meant the Computer Thistlewaite 2-Phase Algorithm.

 

[Daniel Lin]

Hey Guys, I came up with an Idea for doing a 1-look corners(or something that requires both orientation and permutation algorithms) that is very alg-efficient, and could be also taught to noobs as a 2-look. first is to have an OCLL algset that doesn't affect the positions of the corners(like a 2-gen OCLL that restores the positions after) and the regular permutation algorithms. that means that you can 1-look both the orientation and permutation case before you solve them, you just remember, the OLL and the PLL case(and any AUF's in between). This would be useful for methods like corners-first, ortega, roux, pikashit, etc.
recognition would be the same as it is for 1-look corners, i think.

so, what do you think? executing 2 algorithms in 1 look? a good idea?

naw the 2gen pure twists are really slow. If you want CxLL to be one look without learning the algs then just learn how the CO algs affect CP and recognize both steps in one look.

 

[PurpleBanana]

Hey Guys, I came up with an Idea for doing a 1-look corners(or something that requires both orientation and permutation algorithms) that is very alg-efficient, and could be also taught to noobs as a 2-look. first is to have an OCLL algset that doesn't affect the positions of the corners(like a 2-gen OCLL that restores the positions after) and the regular permutation algorithms. that means that you can 1-look both the orientation and permutation case before you solve them, you just remember, the OLL and the PLL case(and any AUF's in between). This would be useful for methods like corners-first, ortega, roux, pikashit, etc.
recognition would be the same as it is for 1-look corners, i think.

so, what do you think? executing 2 algorithms in 1 look? a good idea?

This is similar to an idea I've seen for a 1-look, 2-algorithm OLL system which uses edge-only orientation algs so that the OCLL case can be predicted beforehand.

There is a reason neither is used.

 

[JTWong71]

How is this for a method, Part ZZ and Part Petrus:

I. 2x2x2 Block (In BDL)
II. EO (Steps 1 and 2 can be combined with practice)
III. F2L (Blockbuilding)
IV. Last Layer (EO is done, Subsets such as ZBLL for 1LLL, OCLL/PLL...)

Example Solves:

36 STM
x2 //Inspection
R2 F' R' B' D' R2 L F' //2x2x2 + EO
U2 F2 //EOArrow + 1
R' L' U' R2 //Block
U' l F M' U' R2 //F2L (Cancel into ZZLL)
U2 R2 U' R U' R' U R' U' D R' U R U D' R //ZZLL

37 STM
z2 y //Inspection
D' L' F L B' R F L U' L //EOArrow + 1
U' R D R2 U D' //2x2x3
R U R U R' //F2L
y' R2 U' R2 U' R U2 D' R U' R' U' D R U R2 U2 //ZBLL

 

[wir3sandfir3s]

How is this for a method, Part ZZ and Part Petrus:

I. 2x2x2 Block (In BDL)
II. EO (Steps 1 and 2 can be combined with practice)
III. F2L (Blockbuilding)
IV. Last Layer (EO is done, Subsets such as ZBLL for 1LLL, OCLL/PLL...)

Example Solves:

36 STM
x2 //Inspection
R2 F' R' B' D' R2 L F' //2x2x2 + EO
U2 F2 //EOArrow + 1
R' L' U' R2 //Block
U' l F M' U' R2 //F2L (Cancel into ZZLL)
U2 R2 U' R U' R' U R' U' D R' U R U D' R //ZZLL

37 STM
z2 y //Inspection
D' L' F L B' R F L U' L //EOArrow + 1
U' R D R2 U D' //2x2x3
R U R U R' //F2L
y' R2 U' R2 U' R U2 D' R U' R' U' D R U R2 U2 //ZBLL

Nice idea, I think ZZLL is best for this just because of ZBLL's sheer amount of Algs. It only requires and extra "look" of phasing, which can literally be executed in less than a second with practice and its barely even considered a look.
As far as I can tell, though, it's just Petrus with with an early EO and a restricted block building phase, which might help with speed.
I always thought of ZZ and Petrus as basically the same because Petrus just solves EO later, unlike ZZ, and when you finish The EO in Petrus it basically is ZZ and when F2L is finished for both you get the same results.

 

[JTWong71]

Nice idea, I think ZZLL is best for this just because of ZBLL's sheer amount of Algs. It only requires and extra "look" of phasing, which can literally be executed in less than a second with practice and its barely even considered a look.
As far as I can tell, though, it's just Petrus with with an early EO and a restricted block building phase, which might help with speed.
I always thought of ZZ and Petrus as basically the same because Petrus just solves EO later, unlike ZZ, and when you finish The EO in Petrus it basically is ZZ and when F2L is finished for both you get the same results.

Many people don't like ZZ because of the restrictions in moves early on. With Petrus, people don't use it as a speedsolving method because it requires more thinking and fingertricks are harder, also, no restrictions.

With the method idea, a Petrus block is made, and after EO, you can also manipulate the F-Layer with 180 turns.

 

[wir3sandfir3s]

Many people don't like ZZ because of the restrictions in moves early on. With Petrus, people don't use it as a speedsolving method because it requires more thinking and fingertricks are harder, also, no restrictions.

With the method idea, a Petrus block is made, and after EO, you can also manipulate the F-Layer with 180 turns.

So it's Petrus with feedom of the F face...
This could work. I feel like you can solve most or all of the rest of the 2x2x3 while influencing the LL.

 

[JTWong71]

So it's Petrus with feedom of the F face...
This could work. I feel like you can solve most or all of the rest of the 2x2x3 while influencing the LL.

That is pretty much true as I think about it.
Petrus is already very efficient, but the first two steps don't necessarily have the 'Best' fingertricks.
ZZF2L Statistics show that it takes more average moves than CFOP-F2L.
EO can be done in an average of 3-4 moves.
A 2x2x2 block can be done in around 5-7 moves on average.

I am also trying to figure out a way to influence the last layer during F2L to the point where I can skip the whole LL.