The New Method / Substep / Concept Idea Thread

[IAmAPerson]

New megaminx LSLL:
VHLS
Mega ZBLL

 

[shadowslice e]

New megaminx LSLL:
VHLS
Mega ZBLL

Sorry if I misunderstood, but isn't that something like 1000+ algs?

 

[IAmAPerson]

Sorry if I misunderstood, but isn't that something like 1000+ algs?

Probably lol I don't do mega.

 

[shadowslice e]

5x5x5+ lolmethod
1)Solve as 4x4x4
2) pair/form remaining edges/ centres
3) solves as 3x3x3 (big block corners)

Just the only real practical way I could find to do BBC

Mega lolmethod:
1) Solve a 2x2x2 (3 centres, 4 edges, 1 corner)
2) solve a 2x2x3 (4 centres, 5 edges, 2 corners on the opposite side
3) solve F2L while orienting remaining edges (extend the 2x2x2 to two adjacent centres and merge with the 2x2x3 somewhat so there are only 3 unsolved faces left)*- (solved section now has 9 faces in it)
4) 3-gen S2L
5) LL

*you could orient then do the F2L but this would be slow
This might be an ok method if someone cleverer than me could work out how to orient efficiently.

Petrus big cube method

So if ZZ, Roux and CFOP have their specific big cube methods, why not Petrus? (this is also a bit of a lolmethod so don't expect greatness)

Even cube variant (4x4x4, 6x6x6 etc)

1) build an nxnxn block where n=the size of the cube/2
. 1a) solve 3 adjacent centres
. 1b) solve the edges between them and the corner
2) solve remaining centres
3) pair edges while orienting (like in Z4)
4) 3x3x3 petrus (2x2x2 is already done)

Odd variant

Same but n= (size of cube/2)+1

 

[IAmAPerson]

Petrus big cube method

So if ZZ, Roux and CFOP have their specific big cube methods, why not Petrus? (this is also a bit of a lolmethod so don't expect greatness)

Even cube variant (4x4x4, 6x6x6 etc)

1) build an nxnxn block where n=the size of the cube/2
. 1a) solve 3 adjacent centres
. 1b) solve the edges between them and the corner
2) solve remaining centres
3) pair edges while orienting (like in Z4)
4) 3x3x3 petrus (2x2x2 is already done)

What about this for the Petrus 4x4:

1) Solve centres.
2) Place FLD and BLD corners.
3) Pair up the LD, FD, BD, FL, and BL edges and place them.
4) Pair up and simultaneously orient the other 7 edges.
5) Petrus 2GF2L
6) Petrus LL (OCLL/PLL, COLL/EPLL, etc.)

 

[shadowslice e]

4x4x4->2x2x2

1) build 4 1x2x2s off the four b-corners
2) solve the centres for them with the last edge (either way round)
3) form the corner edge skeletons for the top corners (basically 2x2x2s without the centres)
4) centres cage style
5) 2x2x2

IDK how efficient this is but you could also use this for big block corners on higher order cubes.

 

[Cube Is Life]

I don't usually post ideas here, but I can't remember seeing this before. It's a Skewb method similar to Sarah's intermediate:

1. Solve a layer
2. Solve corners and two opposite sides (L and R or F and B)
3. Finish with L3C: U perm or skip

Would it be any/much worse than Sarah's intermediate? How many cases would there be for step 2?

I'm actually kind of interested in this. If someone could find algs for step 2 that perferably use sledgehammer and hedgeslammer and there are less than like 50 algs/cases I would stop learning Sarah's advanced and start learning this. It seems to me like this would fall in between Sarah's intermediate and Sarah's advanced in terms of difficulty and speed, but who knows maybe it will be super-duper fast.


EDIT: After messing around with my skewb it seems like regognizing the cases would be hard and solving the corners and 2 opposite centers would not be much more efficient than just Sarhah's intermediate. However, this gives me the idea of this method:

Step 1: Layer

Step 2: Solve corners and 1 center (not the center opposite of your layer)

Step 3: Finish. You will either have triple sledge, one move setup to triple sledge, U perm, one move setup to U perm or a skip

 

[IAmAPerson]

A method I thought of a while ago for skewb was this:

Normal variation
1) Solve a layer
2) Solve the opposite center (4 moves)
3) Solve the rest using a subset of Sarah's advanced.

Advanced variation (2x algs)
1) Solve a side (not layer)
2) Solve the opposite center
3) Solve the rest using either a subset of Sarah's advanced or a subset of skewb eg-2.

I would be using this myself if I bothered to learn the algs.

 

[jjone fiffier]

A method I thought of a while ago for skewb was this:

Normal variation
1) Solve a layer
2) Solve the opposite center (4 moves)
3) Solve the rest using a subset of Sarah's advanced.

Advanced variation (2x algs)
1) Solve a side (not layer)
2) Solve the opposite center
3) Solve the rest using either a subset of Sarah's advanced or a subset of skewb eg-2.

I would be using this myself if I bothered to learn the algs.

Funny how everybody thinks advanced is so hard.
If you watch Kennans' tutorial everything is clear.
Btw I think this alredy exsists, 26 algs I think.

 

[theawesomecuber]

3x3 ECL Method

I call this method the "ECL" method, standing for name of the steps.

Steps:

1. EOLine (Just like ZZ)

2. Columns (F2L pairs, then COLL)

3. L6E (Must do Y rotation first, Edges already oriented from EOLine)

Pros:

-If you know ZZ, learning this method is easy
-Little rotations
-Combination of R/U/L and M/U

Cons:

-While building F2L pairs, you have to make sure that no middle edge pieces get "stuck" in DL or DR.

Possible Variation:

A possible variant of this method would be to just do EO, without the line and making sure that no ZZF2L piece gets "stuck" in FD or BD, then do ZZF2L, then COLL, and lastly, L6E.
The benefit of this method is how no Y rotation is needed. However, doing EOLine is marginally slower than just doing EO, so I don't think that doing full ZZF2L in exchange for not having to do a line would be worth it.

 

[shadowslice e]

I call this method the "ECL" method, standing for name of the steps.

Steps:

1. EOLine (Just like ZZ)

2. Columns (F2L pairs, then COLL)

3. L6E (Must do Y rotation first, Edges already oriented from EOLine)

Pros:

-If you know ZZ, learning this method is easy
-Little rotations
-Combination of R/U/L and M/U

Cons:

-While building F2L pairs, you have to make sure that no middle edge pieces get "stuck" in DL or DR.

Possible Variation:

A possible variant of this method would be to just do EO, without the line and making sure that no ZZF2L piece gets "stuck" in FD or BD, then do ZZF2L, then COLL, and lastly, L6E.
The benefit of this method is how no Y rotation is needed. However, doing EOLine is marginally slower than just doing EO, so I don't think that doing full ZZF2L in exchange for not having to do a line would be worth it.

Cool method.

Is this like a ZZ/Roux/PCMS? I did something similar at one point (it was essentially exactly the same as the variant you described) but I founnd that the blockbuilding becomes much more difficult to do quickly as you have a more complex lookahead which is also slower because of restricted moves so I wouldn't go down that path.

The original method is very clever though, combining the blocks with the Eo line. Also, you can still use D,M2/S2 moves to preserve orientation with the pieces so it would not matter too mu h if the piece get "stuck" as they could be fixed quite easily (I use this fact to set up SLS in SSC).

I would therefore list the pros as
1) skip orientation in LSE
2) Ergonomic (<L,U,R,D,M2> and <M,U> for the majority of the solve)
3) Low alg count
4) Easy learning
5) Few rotations.
6) lowish movecount

the cons would be
1) less efficient/restricted block building/pairs
2) slightly more difficult lookahead to the pairs
3) y rotation could be difficult for getting ready for M /moves in the second phase
4) slightly more difficult LSE recognition
But overall, cool method! I think it might be good for you to invest some time to see how fast you can get!

 

[theawesomecuber]

Thanks for responding!

Is this like a ZZ/Roux/PCMS?

Yup, I based it mostly off of ZZ and PCMS, and Roux just happened to work best for L6E.

Also, you can still use D,M2/S2 moves to preserve orientation with the pieces so it would not matter too mu h if the piece get "stuck" as they could be fixed quite easily (I use this fact to set up SLS in SSC).

That's good to know, but it would still be faster to never get the pieces stuck in the first place.

I would therefore list the pros as
1) skip orientation in LSE
2) Ergonomic (<L,U,R,D,M2> and <M,U> for the majority of the solve)
3) Low alg count
4) Easy learning
5) Few rotations.
6) lowish movecount

the cons would be
1) less efficient/restricted block building/pairs
2) slightly more difficult lookahead to the pairs
3) y rotation could be difficult for getting ready for M /moves in the second phase
4) slightly more difficult LSE recognition

For the third con, you could possible do the y rotation right before COLL, so you can have a better transition into L6E. For everything else, I completely agree .

But overall, cool method! I think it might be good for you to invest some time to see how fast you can get!

Thank you! I doubt I'll be spending much time trying to get good at this method, simply for the fact that I'm extremely bad at EOLine. I guess it's something to practice .

 

[shadowslice e]

C4

ok, so I'm going to call this method C4 (partially because it stands for columns 4x4x4 but mostly because it sounds cool ;-p)

It has 2 variants: direct and redux

Direct (sort of cage but not quite and not the more fun of the variants ;-) )

1) solve columns
2) solve F2L using M,m,U and rotations on y
3) solve LL
4) cage style 3rd layer centres.

Redux (2x2x2)

1) create columns composed of 4 pairs on the bottom and 4pairs on the top- they do not have to match.
2) finish 2x2x2 blocks on the bottom
3) create "skeletons" for the 2x2x2 blocks on the top (no centres normally although you could do 1x2x2s with the extra edge) the edges could be done with OELL then PELL which would be 8 OELL and ??? PELL)
4) Centres cage style
5) 2x2x2

 

[IAmAPerson]

Possible Variation:

A possible variant of this method would be to just do EO, without the line and making sure that no ZZF2L piece gets "stuck" in FD or BD, then do ZZF2L, then COLL, and lastly, L6E.
The benefit of this method is how no Y rotation is needed. However, doing EOLine is marginally slower than just doing EO, so I don't think that doing full ZZF2L in exchange for not having to do a line would be worth it.

I thought of this a while ago. The cons were pretty much that the movecount -- which is what makes ZZ so special -- goes down the drain with LSE, even if you're counting in STM instead of HTM. ZZ was made for efficient movecount and fast F2L, and Roux was made for its freeness of blockbuilding and <M, U> 2-gen. Combining these methods doesn't allow either method to reach its full potential. :/ Although this method does sound freaking cool though.

 

[shadowslice e]

Ok, 3x3x3 method

1) 2x2x3 block
2) last cross edge+ one corner
3) orient edges while inserting the last edge
4) Orient the corners using a method of your choice while inserting the last pair.
5) PLL

Meh, nothing too special but it might be a viable method

NOTE: if you like cross, F2L etc, you could just do cross+2 F2L pairs+ 1 corner in any order.


4x4x4+ method (orient first(ish)) completely experimental (so pls dont expect anything amazing)

1) orient the centres (so white/yellow, red/orange and blue/green centres)
2) orient edges either using slices or in the same way as as for zz 3x3x3 by pairing misoriented edges
3) pair edges using double slice or wide moves
4) solve centres cage style
5) ZZ 3x3x3?

NOTE: this is a highly experimental method which will need to be fully checked so don't expect it to be fast especially as some of the concepts are almost completely new as far as I am aware.

 

[mDiPalma]

3x3x3 method

Some advice from a Petrus user:

In step 2, if there ARE misoriented edges, insert a misoriented edge with the corner (in the F2L edge slot) (any edge will work, except the "correct" F2L edge). Half the edges should be misoriented, so this shouldn't take too many extra moves.

Then intuitively orient the rest of the edges and misorient the actual F2L edge that belongs with the previous corner. Leave the misoriented F2L edge in the U-layer.

Then do keyhole to orient the last 2 edges while inserting the F2L edge (exactly 5 moves every time). In my opinion, that will put you at step 4 in the easiest way possible. I actually do this in speedsolves whenever I notice the case.

Also, given that you now have reduced the number of unsolved corners to 5, it might not be too far-fetched to solve CP at this point and finish the solve 2gen.

If you'd like me to demonstrate, please provide a scramble. (I was once falsely accused of giving myself easy scrambles in a method proposal, so I'm scarred for life)

 

[shadowslice e]

Some advice from a Petrus user:

In step 2, if there ARE misoriented edges, insert a misoriented edge with the corner (in the F2L edge slot) (any edge will work, except the "correct" F2L edge). Half the edges should be misoriented, so this shouldn't take too many extra moves.

Then intuitively orient the rest of the edges and misorient the actual F2L edge that belongs with the previous corner. Leave the misoriented F2L edge in the U-layer.

Then do keyhole to orient the last 2 edges while inserting the F2L edge (exactly 5 moves every time). In my opinion, that will put you at step 4 in the easiest way possible. I actually do this in speedsolves whenever I notice the case.

Also, given that you now have reduced the number of unsolved corners to 5, it might not be too far-fetched to solve CP at this point and finish the solve 2gen.

If you'd like me to demonstrate, please provide a scramble. (I was once falsely accused of giving myself easy scrambles in a method proposal, so I'm scarred for life)

Great idea! Thanks

The last 5 corners could likely be done with L5C so that would only leave the 3 ELL cases.
Thus,
1) 2x2x3
2) Cross edge+ 1 corner+ 1 misoriented edge (if there are any)
3) ^^ see above, idk how to describe it quickly
4) L5C
5) ELL

 

[mDiPalma]

Great idea! Thanks

The last 5 corners could likely be done with L5C so that would only leave the 3 ELL cases.
Thus,
1) 2x2x3
2) Cross edge+ 1 corner+ 1 misoriented edge (if there are any)
3) ^^ see above, idk how to describe it quickly
4) L5C
5) ELL

Actually, now I'm thinking that this will only "save" moves if the correct F2L edge is misoriented after the 3x2x2 (but of course, which F2L edge is the "correct" one depends on which corner is easier to solve along with the cross edge....this is getting too complicated). Otherwise, it's better if the inserted "F2L" edge is oriented. I guess you could adjust your approach on the fly to accommodate each case.

And what do you mean L5C? Pair insertion and COLL?

 

[shadowslice e]

And what do you mean L5C? Pair insertion and COLL?

Pretty much, it requires a lot of algs so it might still be more practical to do WV/VHLS/ZBLS etc+ PLL

 

[mDiPalma]

Pretty much, it requires a lot of algs so it might still be more practical to do WV/VHLS/ZBLS etc+ PLL

Have you seen Speed-Heise? I think it would be a good supplement to your methods. In my opinion, I really think it's one of the most efficient 2-step LS/LL approaches possible.

full / easy


Edit: of course, by "most efficient 2-step LS/LL approach," I mean with a reasonable alg count (sorry, ZZLL)