The New Method / Substep / Concept Idea Thread

[Chris_Cube]

Speed Heise and L3C is Snyder-Method in some sense. But snyder never wrote down his algs

 

[PapaSmurf]

What about Speed-Heise > algorithmic L3C? I've heard L3C is one of the best ZBLL substeps, and Speed-Heise isn't that bad.

(Sorry about all the is x or y better questions)

Yeah, speed heise isn't that bad, but for the majority of cases, full ZBLL is better (also, slot neutrality is important too).

 

[Username: Username:]

Yeah, speed heise isn't that bad, but for the majority of cases, full ZBLL is better (also, slot neutrality is important too).

what is slot neutrality?

I'm pretty sure it means you can do the alg for any slot.

So you mean, slot neutrality means that you can do like, back slot, diagonal slot and front slot comfortably right?

 

[brododragon]

what is slot neutrality?

I'm pretty sure it means you can do the alg for any slot.

 

[PapaSmurf]

So, for insert->ZBLL, you can do that for any slot. The more you influence LL in LS, the less likely you'll be able to be slot neutral (take the extreme example of L5C->L5EP, you'd either have to be really good at mirroring algs that are bad from BL, or you just solve to FR every time, which makes your F2L worse).

 

[The Pocket Cuber]

Alright so I’ve been experimenting with the PCMS method for solving a 3x3 (link below), and I noticed that the first steps for building columns are great, but realised that the Last eight edges step needs development. And with experimentation for a week, I now average 22.5 seconds with this version of PCMS and believe it could be a viable speedcubing method for years to come. It averages well under 50 moves, whilst speedcubing, and has extremely good lookahead. Without further ado, here’s what I propose

 

[ProStar]

Alright so I’ve been experimenting with the PCMS method for solving a 3x3 (link below), and I noticed that the first steps for building columns are great, but realised that the Last eight edges step needs development. And with experimentation for a week, I now average 22.5 seconds with this version of PCMS and believe it could be a viable speedcubing method for years to come. Without further ado, here’s what I propose

This is similar to what me and @WarriorCatCuber came up with for L8E:

EO
DL+DR
UL+UR
Roux L4E

 

[The Pocket Cuber]

This is similar to what me and @WarriorCatCuber came up with for L8E:

EO
DL+DR
UL+UR
Roux L4E

Yes I did briefly mention that during the document, though, for many reasons listed in the document, is slightly worse than my proposed method. Though it is a viable way to approach it especially since no Algs.

Edit: It would be greatly appreciated if you guys let me know whether the 8 example solves work and whether they explain the method. I think I’ve provided enough information as to why this could be a seriously top method throughout the document too.

 

[brododragon]

Here's an idea I've been thinking about:
1. EOCross
2. Sort remaining edges into the correct layer/slice
3. Orient First Layer Corners
4. Permute first layer corners and E-slice edges
5. ZBLL

Phasing (ZZLL) and other strategies can be used to significantly decrease move-count. For example, you could solve the E-slice or U-layer edges to get a band with EO and solved D-layer for an easy finish or COLL, respectively. Actually, now that I think about it, the COLL version would be great.

 

[TheRouxGuy]

@brododragon How many moves on an average?

 

[The Pocket Cuber]

Here's an idea I've been thinking about:
1. EOCross
2. Sort remaining edges into the correct layer/slice
3. Orient First Layer Corners
4. Permute first layer corners and E-slice edges
5. ZBLL

Phasing (ZZLL) and other strategies can be used to significantly decrease move-count. For example, you could solve the E-slice or U-layer edges to get a band with EO and solved D-layer for an easy finish or COLL, respectively. Actually, now that I think about it, the COLL version would be great.

The only problem I can see in this method is how on earth are you going to orient the first layer corners. If it is in a way we’re you kind of do F2L with the E layer edges, then I think CFOP would outdo it. I just don’t think it would be easy at all to orient the first layer corners without knocking out an e layer and edg or stuffing up orientation of edges.

 

[PapaSmurf]

Here's an idea I've been thinking about:
1. EOCross
2. Sort remaining edges into the correct layer/slice
3. Orient First Layer Corners
4. Permute first layer corners and E-slice edges
5. ZBLL

Phasing (ZZLL) and other strategies can be used to significantly decrease move-count. For example, you could solve the E-slice or U-layer edges to get a band with EO and solved D-layer for an easy finish or COLL, respectively. Actually, now that I think about it, the COLL version would be great.

Ngl, this sounds like a much worse ZZ. EOCross F2L is super good and there are much better ways to solve it than that. Also, phasing doesn't decrease the movecount, it increases it. It only reduces the alg count (but you're better off just learning ZBLL anyway).

 

[brododragon]

Ngl, this sounds like a much worse ZZ. EOCross F2L is super good and there are much better ways to solve it than that. Also, phasing doesn't decrease the movecount, it increases it. It only reduces the alg count (but you're better off just learning ZBLL anyway).

Oops I meant alg count. how does the COLL/EPLL compare?

@brododragon How many moves on an average?

I don't know; Some algs need genning. I'll open up Cube Explorer and gen algs as I need them to see the move-count.

The only problem I can see in this method is how on earth are you going to orient the first layer corners. If it is in a way we’re you kind of do F2L with the E layer edges, then I think CFOP would outdo it. I just don’t think it would be easy at all to orient the first layer corners without knocking out an e layer and edg or stuffing up orientation of edges.

It would be just like ZZ EO except whenever you do an R/L move you would have to undo it. I tested it out a few times and it was fine.

Also, one advantage to this is that it requires very little skill. Roux has F2B, ZZ has ZZF2L and EO, Petrus has F2L and EO, and CFOP has F2L. All you have to understand in this method is EOCross (there are good tutorials) and sorting edges (Ryan Heise has a great tutorial on it). This makes it extremely easy to understand the method and get to it's full potential quickly.

 

[PapaSmurf]

OCLL/PLL is 28 algs and is faster than COLL/EPLL for TH. Anyway, I still think that normal F2L would be faster.

 

[slowhandzboi]

im proposing an algset

Called L5C which solves the Final f2l corner
and The Last layer corner's orientationa and permutation
which leaves you with an ELL case

 

[AlphaCuber is awesome]

im proposing an algset

Called L5C which solves the Final f2l corner
and The Last layer corner's orientationa and permutation
which leaves you with an ELL case

pretty sure this already exists

 

[slowhandzboi]

the only downside is the number of algs which range about 500-600

im proposing an algset

Called L5C which solves the Final f2l corner
and The Last layer corner's orientationa and permutation
which leaves you with an ELL case

pretty sure this already exists

you sure about that

if youre gonna say CLS


CLS only solves the orientation

and if youre gonna say CPLS

CPLS only solves last layer corner permutation and not orientation

 

[Hazel]

im proposing an algset

Called L5C which solves the Final f2l corner
and The Last layer corner's orientationa and permutation
which leaves you with an ELL case

Another member posted this same idea before. It's generally regarded to be not worth learning, as solving LS normally followed by regular last layer is just faster. ELL isn't a great set either, some of the recognition is really iffy.

 

[slowhandzboi]

Another member posted this same idea before. It's generally regarded to be not worth learning, as solving LS normally followed by regular last layer is just faster. ELL isn't a great set either, some of the recognition is really iffy.

aw man

 

[ProStar]

Best way to do L5C would probably be with comms anyway because of the alg count. Also, please remember to search the forums before posting a thread, as often there will be a thread meant for such topics. For example, here is the thread for new method/concept/substep proposals.