The New Method / Substep / Concept Idea Thread

[Neuro]

In L7E, why don't we insert LE while doing EO and then 4B/4C? It should give decent movecounts w/o too many algs. And if you want to go the extra mile, implement L6EP.

I think if you use L7E w/ 42/TCMLL neutrality you *may* get a sub-40 avg movecount with 274 algs (unless you go for algorithmic L6EP, then it goes up significantly). I'm worried about L5C recog but if it can be absolutely nailed then this has amazing potential.

EXAMPLES (AVG=17.2)
16
16
19
18
17

EDIT: I think I'm going to gen a ton of algs and compile all CP Roux based methods that have merit and make a thread. So far I have mine (2GRoux, uses Pinkie Pie) the L5C+LE variant, a L7E variant, the base (uses CMLL/LSE), this unnamed method (CPFB, <RrMU> EO Line, F2L, 2GLL), and one similar to LLOB+ZBLL (CPFB, SB, EODFDB, 2GLL) I'll also include some LSLL variants as necessary but I'd like to give credit to the OG creators of these methods so please respond if you've already invented one of these variants. Thanks!

 

[Thermex]

In L7E, why don't we insert LE while doing EO and then 4B/4C? It should give decent movecounts w/o too many algs. And if you want to go the extra mile, implement L6EP.

I think if you use L7E w/ 42/TCMLL neutrality you *may* get a sub-40 avg movecount with 274 algs (unless you go for algorithmic L6EP, then it goes up significantly). I'm worried about L5C recog but if it can be absolutely nailed then this has amazing potential.

EXAMPLES (AVG=17.2)
16
16
19
18
17

The L7E method you proposed is pretty similar to @crafto22's but with a slight boost in efficiency due to the first two step being combined. I'm honestly not sure if it's worth it for the extra algs, but it can be two-looked which is a bonus. Anyway here's a pretty cool way to implement L7E, it's nice and simple, "potentially" sub-40 moves, and pretty ergonomic:

1.) FB+BRSquare (15/15)
2.) Tripod L4E (place one oriented corner at UBR and solve the rest of the corners in one alg) (8/24)
3. L7E (17/40?)

About 100 algs for this; really good but a supieror L7E method still needs to be devoloped. I honestly think we should all just dedicate our cubing time to making a solid L7E method, it's definitely one of the most important developments yet to be made. Now that I think about it, it might be better to switch steps 2 and 3 of my method, making this:

1. FB+BRSquare (15/15)
2. Insert FR edge while orienting remaining edges (<r, U> based and fast, ~7/22 moves?)
3. Tripod L4E (10/32)
4. PL6E (like @Neuro said, this could be done algorithmic-ly or intuitely, either way it's around 7-9 moves/40)

This is also about 40 moves, but with an <r, U> FR+EO stage that's much more ergonomic than M, F, and U moves. I think I'll spend most of tomorrow genning the L4E cases I was talking about and update you guys on how this method turns out. It looks like it has a lot of potential

 

[TDM]

1.) FB+BRSquare (15/15)
2.) Tripod L4E (place one oriented corner at UBR and solve the rest of the corners in one alg) (8/24)
3. L7E (17/40?)

I think the movecount is a bit closer to 42

 

[Neuro]

My L7E method would have a minimum of 30 algs, but you could go up to 60 or even 120. 30 can solve only FR when LE is in a constant spot. 60 algs solves either FR or BR. 120 allows the edge to be in either DF or DB. This assists in recognition by always having oriented centers. Still, I'd say that 60 algs (same as EOLR) is very good and quite feasible for one to learn.

 

[shadowslice e]

My L7E method would have a minimum of 30 algs, but you could go up to 60 or even 120. 30 can solve only FR when LE is in a constant spot. 60 algs solves either FR or BR. 120 allows the edge to be in either DF or DB. This assists in recognition by always having oriented centers. Still, I'd say that 60 algs (same as EOLR) is very good and quite feasible for one to learn.

You could just do it intuitively like @Teoidus and I already do.

 

[Thermex]

I think the movecount is a bit closer to 42

I just realized this is really similar to 42, but there are still differences. The main thing is that instead of doing a conjugated CMLL, which in my expericnece has very difficult recog, you learn a set of 96 algs that solve the 4 pieces without the ARF. Small difference, but it could help quite a lot as I predict the Tripod L4E cases are pretty fast.

My L7E method would have a minimum of 30 algs, but you could go up to 60 or even 120. 30 can solve only FR when LE is in a constant spot. 60 algs solves either FR or BR. 120 allows the edge to be in either DF or DB. This assists in recognition by always having oriented centers. Still, I'd say that 60 algs (same as EOLR) is very good and quite feasible for one to learn.

With AUFS I estimate your method to be about 100 algs (14 orientations×7 spots the FR edge can go in=98, plus the cases with EO done is about 100). This is about the same as Crafo's method, although some cases might be cut off by rotational symmetry. Today I think I'll do some tests solves with both methods and tell you which approach I found faster.

You could just do it intuitively like @Teoidus and I already do.

How do you do L7E intuitively? Is it anywhere near as good as some of the algorithmic approaches?

 

[Teoidus]

I just realized this is really similar to 42, but there are still differences. The main thing is that instead of doing a conjugated CMLL, which in my expericnece has very difficult recog, you learn a set of 96 algs that solve the 4 pieces without the ARF. Small difference, but it could help quite a lot as I predict the Tripod L4E cases are pretty fast.

inb4 all the algs in this algset look quite a bit like CMLL conjugates

How do you do L7E intuitively? Is it anywhere near as good as some of the algorithmic approaches?

yes

 

[Teoidus]

wait a minute... why is it called Tripod L4E???? A tripod is not formed anywhere, corners are being solved (not edges) and there are five of them (not four).

In fact here is a color-coding of your naming scheme (green = accurately describes step, red = inaccurately describes step):
Tripod L4E

 

[shadowslice e]

... conjugated CMLL, which in my expericnece has very difficult recog

I don't know what you're doing for recog but in my experience it's only very slightly harder than CMLL and easier than ZBLL.

...you learn a set of 96 algs that solve the 4 pieces without the ARF.

You shouldn't have to ARF in order to solve. Ideally you would just build the block so it ends up that way. In more advanced versions you don't even have to orient the corner.

Small difference, but it could help quite a lot as I predict the Tripod L4E cases are pretty fast.

The BTR algs are likey faster (especially when set specific algs are used instead of just CMLL) because I predict most of the algs for "L4E" will start with an R which BTR avoids (and most BTR algs don't start with an R')

While we're here, I'll just list all the reasons for BTR over L4E:
1) The algs are shorter (explained above)
2) Recog is not harder (because you look at the same number of stickers and have the same number of cases)
3) It sets up nicely for intuitive L7E
4) There are less algs
5) It can be used more easily with other advanced techniques such as NMLL.

Also, why are you calling it L4E? You don't solve any edges let alone the last 4.

How do you do L7E intutively? Is it anywhere near as good as some of the algorithmic approaches?

The movecount is the same. I average about 16 moves with it without any algs and it also becomes much more possible to force nice cases because you are not restricted in that sense.

 

[Teoidus]

While we're here, I'll just list all the reasons for BTR over L4E:
1) The algs are shorter (explained above)
2) Recog is not harder (because you look at the same number of stickers and have the same number of cases)
3) It sets up nicely for intuitive L7E
4) There are less algs
5) It can be used more easily with other advanced techniques such as NMLL.

6) It solves the corners and not the edges like it's supposed to

 

[mDiPalma]

Hello everyone. How are you?

There are too many methods on this thread, and i have no idea what's going on. let's thin them out.

any method that cant get "sub-40 htm" on this scramble shall be permanently eliminated from discussion because it's obsolete (mods WILL enforce this with permabans!). please provide a linear solution to the following scramble before you continue blabbering.

B2 R2 B' D2 L2 F2 L2 F D2 R2 U B L' R F R U' L' B2 D'

thank you for your time, and god bless america

Spoiler: petrus method (28 htm, 27 stm)

y' x' B' U' F' R U' F B' L
F R F' U R' U2 F' U F R U // R' cancels
R2 F2 R' B2 R F2 R' B2 R2

Spoiler: zz method (29 htm, 27 stm)

D F D' R' B L U F'
R' L U' L2 U' R L' U' L2
U2 R' U R2 U2 R' U R2 U2 R2 U' R

Ranking	Method	Movecount	Submitter
1.	Petrus	28 htm	mdipalma
2.	ZZ	29 htm	mdipalma
3.	2GR	35 stm	elo13
3.	cpfl->2gll	35 stm	elo13
5.	Zzoux	38 htm	pyjam
5.	CFOP	38 htm	elo13
7.	Briggs	39 htm	arc

 

[hellomyfriend]

New ZZ variant

Okay with the research I have done I don't think anyone thought of this. Tell me if some one did. It is called ZZ-ST or ZZ-Shower Thoughts. It is a 2-3 Look Last Block Last Layer depending on variant. So This is how it goes.

CBeLL Variant

1. EOline- So just solve EOLine.
2. Solve the first square of the 1st block in ZZ2L
3. While finishing the first block Solve the 2nd block corners.
4. Do CBeLL - This is a COLL variant which affects second block edges to allow more efficient algs.
5. L7EP - Last Seven Edges Permutation. You can two look this for 3LLBLL or one look for 2LLBLL.

CBLL Variant
Here instead if solving second block corners in the 3rd step you can just insert them without worrying about them being completely solved but, now there is 6x the number of cases in the 4th step.

So if anyone can help with the math since I actually don't know how to calculate the movecount for individual steps.

 

[shadowslice e]

Th

New ZZ variant

Okay with the research I have done I don't think anyone thought of this. Tell me if some one did. It is called ZZ-ST or ZZ-Shower Thoughts. It is a 2-3 Look Last Block Last Layer depending on variant. So This is how it goes.

CBeLL Variant

1. EOline- So just solve EOLine.
2. Solve the first square of the 1st block in ZZ2L
3. While finishing the first block Solve the 2nd block corners.
4. Do CBeLL - This is a COLL variant which affects second block edges to allow more efficient algs.
5. L7EP - Last Seven Edges Permutation. You can two look this for 3LLBLL or one look for 2LLBLL.

CBLL Variant
Here instead if solving second block corners in the 3rd step you can just insert them without worrying about them being completely solved but, now there is 6x the number of cases in the 4th step.

So if anyone can help with the math since I actually don't know how to calculate the movecount for individual steps.

Can anyone create an example solve with this scramble so this method doesn't get deleted.
B2 R2 B' D2 L2 F2 L2 F D2 R2 U B L' R F R U' L' B2 D'

This is an interesting method which seems like it could work fairly well with LMCF. My main query is why you've broken up the steps in the way you have? Why not just do EOLine, FB, 2 corners, COLL (with variant algs for nicer EP no real need to call it a separate alg set)?

That said I think hinges very similar to it have been proposed though nothing identical so it might still be worth exploring (though it may be that there are some inefficiencies caused by solving edges after corners as the corners get "in the way").

With that said I wish you luck if you continue exploring this variant!

 

[hellomyfriend]

Th

This is an interesting method which seems like it could work fairly well with LMCF. My main query is why you've broken up the steps in the way you have? Why not just do EOLine, FB, 2 corners, COLL (with variant algs for nicer EP no real need to call it a separate alg set)?

That said I think hinges very similar to it have been proposed though nothing identical so it might still be worth exploring (though it may be that there are some inefficiencies caused by solving edges after corners as the corners get "in the way").

With that said I wish you luck if you continue exploring this variant!

What is LMCF?
Also I broke it up like this becuase this was how it originally formed in my head.

 

[shadowslice e]

What is LMCF?

It's somehing that @efattah has been working on and it's basically an algorithmic version of corners first solving so I think you might be interesting in the approach to solving the edges.

 

[hellomyfriend]

It's somehing that @efattah has been working on and it's basically an algorithmic version of corners first solving so I think you might be interesting in the approach to solving the edges.

Any links to it?

 

[shadowslice e]

Any links to it?

Original thread
There's also a wiki page on it iirc.

E: you might also like looking up waterman on the wiki but as LMCF is almost Waterman+algs it might not be so much benefit.

 

[Pyjam]

B2 R2 B' D2 L2 F2 L2 F D2 R2 U B L' R F R U' L' B2 D'

Zzoux : 39 STM (40 HTM) – OK ?

Or: Zzoux : 38 STM (38 HTM)

zz D' R2 U2 L2 F u' x' // (6) Left block
R' U r' U2 r' U r U' r2 // (9/15) EO-line
R U2 R U2 R2 U R' U' R // (9/24) Right block
y' R2' F2 R U2 R U2' R' F2 R U' R' U R U // (14/38) ZBLL T

 

[Abram Lookadoo]

any method that cant get "sub-40 htm" on this scramble shall be permanently eliminated from discussion because it's obsolete (mods WILL enforce this with permabans!). please provide a linear solution to the following scramble before you continue blabbering.

B2 R2 B' D2 L2 F2 L2 F D2 R2 U B L' R F R U' L' B2 D'

I'm not sure what power you have, but i feel this is a bit rude. why would you demote bad to kind of good methods, even if they provide useful information to this thread? and if this is a thing that, for some reason starts, you should probably put, like, five scrambles, each favoring diferent starts. you would have to complete 3/5 scrambles in less than 45 moves in snyder metric. so, if you would, put more thoughts into the scrambles.

edit: the 5 scramble thing is not as good as an idea as I've thought. instead read the first post of this thread, you'll find it important. but, until you get a premium member's approval along with the active majority's approval, this rule will not take effect.

 

[Teoidus]

To be honest I don't think it's too rude. It's a very sassy way of pointing out how some have been throwing the "sub-40 linear" label around very liberally (a pointing out that is, in my opinion, very well deserved).