The New Method / Substep / Concept Idea Thread

[BenChristman1]

That's been thought of by most people. It's really not that good.

No, its not.

 

[Cubingcubecuber]

No, its not.

WombatWarrior17 is right

 

[ImmolatedMarmoset]

I came up with this weird method where you solve 2 1x2x3 blocks, just like Roux, but then using the M slice, you insert the other 2 cross pieces, then solve the last layer like normal.

No, its not.

Very common, yeah. The only good variant is ZBRoux, which uses ZBLL. The problem is that it’s so many more algs for a small improvement. So many people have come up with it, we get it like once every one or two weeks here.

 

[Cubinwitdapizza]

F2L-1-1cross piece next to the pair/2x2x3block+pair
EO+LS(could probably do both at once, but doing them separately isn’t too bad)
COLL(or maybe do last pair,
L5EP

Seems pretty decent to me

I think this is petrus with EJLS? And then L5EP. I believe this was proposed recently.

 

[Cubingcubecuber]

I think this is petrus with EJLS? And then L5EP. I believe this was proposed recently.

That’s not what EJLS is

 

[Cubinwitdapizza]

That’s not what EJLS is

Oh sorry. I believe petrusquber asked if he should do this kinda thing earlier.

 

[Cubingcubecuber]

Is there a last layer subset in which the corners are oriented, and than you solve the cube(eo+cp+ep) Like ELL, but with CP. It would be 174 algorithms, -21 PLL, -25 ELL = 128 This could be used with WV as a low(er than ZB) alg count 2 look LS+LL method. If this is new, I will name it KLMNLL

 

[PapaSmurf]

That's not new. COALL has been around for a while (also called ELLCP or CPELL) and isn't better than ZB, Zipper, Ribbon with CLS (which gives the COALL ending).

 

[brododragon]

This idea is super undeveloped and I hope someone can help.

First, solve the first layer. Don't know how, but just do it.

Next, do the second layer, but while doing that, find bad edges and corners and place them in certain places so that they naturally get fixed my the second layer.

Finally, permute the LL.

Like I said, super undeveloped. Help developing it would be greatly appreciated.

 

[Etotheipi]

This idea is super undeveloped and I hope someone can help.

First, solve the first layer. Don't know how, but just do it.

Next, do the second layer, but while doing that, find bad edges and corners and place them in certain places so that they naturally get fixed my the second layer.

Finally, permute the LL.

Like I said, super undeveloped. Help developing it would be greatly appreciated.

Interesting, I suppose one could develop multiple edge insertion algs that orient specific pieces, and then apply the correct ones.

From fiddling, I found that the alg U R U' R' U' F' U F orients two corners and one last layer edge, so you could manipulate it to solve EO every time, but for the corners I think we'll need another alg.

 

[Etotheipi]

This is mostly waterman. In waterman, you solve first layer, than solve E layer, and then last layer. The E layer is typically done with commutators. I guess your method is different, in the way that you kinda do like a domino sort of thing.

I think this method is LBL, but with an forced OLL skip each time.

 

[Cubinwitdapizza]

I’m dumb, waterman is you solve first layer and then do a CxLL, then solve r edges, whilst orienting m layer edges, and then, L4e.

 

[Etotheipi]

I’m dumb, waterman is you solve first layer and then do a CxLL, then solve r edges, whilst orienting m layer edges, and then, L4e.

Ok then. =P

 

[PapaSmurf]

There's a set of methods called L2L. Check them out.

 

[brododragon]

Interesting, I suppose one could develop multiple edge insertion algs that orient specific pieces, and then apply the correct ones.

From fiddling, I found that the alg U R U' R' U' F' U F orients two corners and one last layer edge, so you could manipulate it to solve EO every time, but for the corners I think we'll need another alg.

I do not have access to a cube right now, but have two ideas for the corners.

Idea: Build the white side middle first. Then do the right side. You can use the left to orient the corners before actually constructing the left side. Finally, permute.

Edit: it's just dawning on me that this could be a good OH method; mostly R and U, with a couple Fs.

 

[brododragon]

There's a set of methods called L2L. Check them out.

What I'm trying to do is to orient edges and corners during the first layer

Edit: Whoops I got confused

 

[brododragon]

Interesting, I suppose one could develop multiple edge insertion algs that orient specific pieces, and then apply the correct ones.

From fiddling, I found that the alg U R U' R' U' F' U F orients two corners and one last layer edge, so you could manipulate it to solve EO every time, but for the corners I think we'll need another alg.

Wait, so you use U' R' U' F' U F to solve EO, right? Also, where is the other edge that gets flipped? (two edges always get flipped).

 

[Etotheipi]

Wait, so you use U' R' U' F' U F to solve EO, right? Also, where is the other edge that gets flipped? (two edges always get flipped).

the other edge that gets oriented is the one being inserted.

 

[brododragon]

Interesting, I suppose one could develop multiple edge insertion algs that orient specific pieces, and then apply the correct ones.

From fiddling, I found that the alg U R U' R' U' F' U F orients two corners and one last layer edge, so you could manipulate it to solve EO every time, but for the corners I think we'll need another alg.

Couldn't you use a slightly modified alg to insert a white edge to the bottom layer whilst orienting two LL corners? You could do this twice to make a white cross AND orient all corners. To preserve them, it shouldn't be more than a move or two inserted into the middle of algorithm.

Note: I've been packing for a trip for the last few hours, so sorry about the delay.

Edit: Do you want to continue the conversation in an actual conversation?

 

[Etotheipi]

Couldn't you use a slightly modified alg to insert a white edge to the bottom layer whilst orienting two LL corners? You could do this twice to make a white cross AND orient all corners. To preserve them, it shouldn't be more than a move or two inserted into the middle of algorithm.

Note: I've been packing for a trip for the last few hours, so sorry about the delay.

Edit: Do you want to continue the conversation in an actual conversation?

Won't the first layer be solved by the time you start on orienting things? I guess you could do that though.

I go on long car drives every month or so, so no worries about packing, I'm used to it lol. (I go somewhere for Christmas on the 20th so I won't be active at all that day, except maybe for a little while around 8:00 or 9:00.)