The New Method / Substep / Concept Idea Thread

[JTWong71]

thats what it was, yes. Seems like a better Idea to teach than LBL.

I also had an Idea where you solve the D layer, solve U corners, then use setup moves and the H-perm(or T-perm if you want to reuse algs) to solve all the rest of the edges. pretty much a mix of LBL and old pochmann(which should totally be a beginners method BTW)

I made an Example Solve with it, and the only problem I have with for teaching beginners are for Edge Parity and Edge Orientation to solve an Edge completely.

y //Inspection
R2 U F' R' F2 R2 //D-Layer Corners
(B D' B') M2 (F L F') U2 M2 //Finish First Layer
U R U R' U R U' R D R' U' R D' R2 //Solve Corners
(F) M2 U M2 U2 M2 U M2 (F') //Edge-1
(d L') M2 U M2 U2 M2 U M2 (L) //Edge-2
(d' R) M2 U M2 U2 M2 U M2 R' //Edge-3
(U' R') M2 U M2 U2 M2 U M2 (R)//F2L
R U R' U' R' F R2 U' R' U' R U R' F' //LL Edge-1
(M D' L2) R U R' U' R' F R2 U' R' U' R U R' F' (L2 D M') //LL Edge-2
M' U' M' U' M' U2 M U' M U' M U2 //Parity

 

[gyroninja]

What he never mentioned anything about the edges being phased or not, he just said to solve them. How is he suggesting a worse form of phasing?

He's phasing the edges adj instead of opposite.

 

[crafto22]

He's phasing the edges adj instead of opposite.

Oh I see what you mean by leaving two edges out he's technically phasing them.

 

[Sion]

Here's my shot at what I would consider to be what I would've came up with if I were an 80's cuber, as a method of solving (It's something I came up with myself completely, and I just need to learn how to create algs for it)

-solve one cross on one side, and 3/4 cross on the other side intuitively
- use a permutation alg to complete the 3/4 cross
- solve the remaining four corner-edge-corner pairs in four looks, solving one column at a time

I find this method to be more general in it's steps and solving procedures, but generating algs for it will be very tedious!

 

[wir3sandfir3s]

Here's my shot at what I would consider to be what I would've came up with if I were an 80's cuber, as a method of solving (It's something I came up with myself completely, and I just need to learn how to create algs for it)

-solve one cross on one side, and 3/4 cross on the other side intuitively
- use a permutation alg to complete the 3/4 cross
- solve the remaining four corner-edge-corner pairs in four looks, solving one column at a time

I find this method to be more general in it's steps and solving procedures, but generating algs for it will be very tedious!

So basically reverse PCMS (MSPC?)... Algs for last edge shouldn't even be needed. Tbh, no matter how you solve the columns, Roux will be more efficient. For example, say you solve the columns by doing all corners in 2 looks then edges, you will be left with L4E which is best solved on M slice, but here it will be E slice. So you tilt the whole method by a Z move. Basically you are just creating the Roux blocks restrictively and inefficiently and adding 2 extra edges which you solve in L6E... Definitely an 80's idea, but not that great...

 

[OLLiver]

Has anyone thought of combining Roux+ZB?
Because Roux has a low movecount, how about solving F2B, orient all edges, permuting the bottom two, leaving a ZBLL case on top.
This would have a low move count and recog for ZBLL isn't much harder than CMLL. Unsure about move count for orienting all edges and permuting bottom edges.
Thoughts?

 

[TDM]

Has anyone thought of combining Roux+ZB?
Because Roux has a low movecount, how about solving F2B, orient all edges, permuting the bottom two, leaving a ZBLL case on top.
This would have a low move count and recog for ZBLL isn't much harder than CMLL. Unsure about move count for orienting all edges and permuting bottom edges.
Thoughts?

Yes, people have thought of this before. Permuting the bottom two edges would be more difficult than ULUR, so you're really asking if CMLL+L4E is faster than ZBLL: and I don't think it is. Lookahead is also worse. So I'm fairly sure Roux is faster.

But you could do it if you really liked ZBLL

 

[4Chan]

Has anyone thought of combining Roux+ZB?
Because Roux has a low movecount, how about solving F2B, orient all edges, permuting the bottom two, leaving a ZBLL case on top.
This would have a low move count and recog for ZBLL isn't much harder than CMLL. Unsure about move count for orienting all edges and permuting bottom edges.
Thoughts?

I once thought about doing ZBLL instead of CMLL to influence the edge permutation, but I wasn't good enough at it. ):

 

[Shiv3r]

Yes, people have thought of this before. Permuting the bottom two edges would be more difficult than ULUR, so you're really asking if CMLL+L4E is faster than ZBLL: and I don't think it is. Lookahead is also worse. So I'm fairly sure Roux is faster.

But you could do it if you really liked ZBLL

I had an idea: why not solve F2B and EO, then place LR edges(easier to track), and then do U/U' M2 and execute a ZBLL(probably a 21LL/EZBLL for all them lazy folks-- ~45 algs for all them cases), then undo setups.

 

[Shiv3r]

the reason I hate CFOP is the rotations, so i was playing with a rotationless F2L today. I call it ROTFL (Rotationless F2l/Fridrich Last Layer). its sorta like blockbuilding a freeFOP, except for one thing: no rotations.
so here's how i did it:
Roux FB
keyhole in one or both D edges(sometimes left em just because it was easier to leave em out)
finish F2L like petrus
OLL/PLL

after a few tries, I almost got a sub-20 time(I avg about 25-27 secs). the surprising thing: my TPS was crappier than my normal roux TPS, yet my times were faster.

 

[Sion]

the reason I hate CFOP is the rotations, so i was playing with a rotationless F2L today. I call it ROTFL (Rotationless F2l/Fridrich Last Layer). its sorta like blockbuilding a freeFOP, except for one thing: no rotations.
so here's how i did it:
Roux FB
keyhole in one or both D edges(sometimes left em just because it was easier to leave em out)
finish F2L like petrus
OLL/PLL

after a few tries, I almost got a sub-20 time(I avg about 25-27 secs). the surprising thing: my TPS was crappier than my normal roux TPS, yet my times were faster.

That's sort-of another form of FreeFOP using F2B for F2L. I tried it, I suggest learning ZZ, since using it feels like CFOP, but ZZ F2L has no rotations.

 

[Shiv3r]

That's sort-of another form of FreeFOP using F2B for F2L. I tried it, I suggest learning ZZ, since using it feels like CFOP, but ZZ F2L has no rotations.

I use roux. Eoline is hard, i average about 15 seconds more with ZZ and thats just the EOline.
I also like roux a lot better. but the idea still stands, and although I have llower TPS than I normally would for CFOP, I still get at least 8 seconds better on average over CFOP.

 

[Shiv3r]

That's sort-of another form of FreeFOP using F2B for F2L. I tried it, I suggest learning ZZ, since using it feels like CFOP, but ZZ F2L has no rotations.

and its not really F2B->F2L, its more like 1) FB 2)solve the rest of F2L with r, M, U moves

 

[Sion]

and its not really F2B->F2L, its more like 1) FB 2)solve the rest of F2L with r, M, U moves

I did try mixing blockbuilding styles with Orienting/Permuting styles, better left separate. It seems faster since OLL/PLL isn't intuitive like something as LSE, but while you get quicker in that manner, someone using LSE will get faster since they have two less cubies to permute and once they get killer recognition, they can glide through it with a faster TPS.

I'm not saying your method is terrible, but i'm thinking you should try another last step that's unique to this method that could be faster, maybe something such as having the last CE pair Permute the remaining corners, sorta like a winter variation, which can Lead to something like an ELL, LSE, or L5EOP/EPLL combo.

those are some of my ideas thrown into the ring.

 

[TDM]

I had an idea: why not solve F2B and EO, then place LR edges(easier to track), and then do U/U' M2 and execute a ZBLL(probably a 21LL/EZBLL for all them lazy folks-- ~45 algs for all them cases), then undo setups.

This actually sounds like it could be good. The ZBLL recognition will be very difficult and you're still using more moves than Roux, so I don't think it's faster - but if you can find a good recognition system I think it'll be a good method.

(unless there's something obvious I'm missing here)

 

[Sion]

This actually sounds like it could be good. The ZBLL recognition will be very difficult and you're still using more moves than Roux, so I don't think it's faster - but if you can find a good recognition system I think it'll be a good method.

(unless there's something obvious I'm missing here)

i'm thinking for this maybe you can use a winter variation CLL with the last pair, and it could potentially lower movecount, like L5C&P could maybe be a name for that substep.

Say, thinking about it, it actually may help with roux itself once developed!!

 

[bobthegiraffemonkey]

This actually sounds like it could be good. The ZBLL recognition will be very difficult and you're still using more moves than Roux, so I don't think it's faster - but if you can find a good recognition system I think it'll be a good method.

(unless there's something obvious I'm missing here)

To clarify:
1) Solve UL/UR or UF/UB directly to DF/DB
2) ZBLL, using the fact that two of the LL edges have a different colour on top, maybe that could actually be useful?

Hmm, sounds interesting, I hope someone looks into it (I don't do Roux or ZBLL).

 

[OLLiver]

To clarify:
1) Solve UL/UR or UF/UB directly to DF/DB
2) ZBLL, using the fact that two of the LL edges have a different colour on top, maybe that could actually be useful?

Hmm, sounds interesting, I hope someone looks into it (I don't do Roux or ZBLL).

I can do ZBLL but my roux skills are limited

 

[shadowslice e]

I can do ZBLL but my roux skills are limited

I can do roux but my ZBLL skills are limited

 

[OLLiver]

I can do roux but my ZBLL skills are limited

well this sucks.