The New Method / Substep / Concept Idea Thread

Small Poster said:

Petrus? What is that?

Click to expand...

A very similar method. However, to make the rest of the solve 2 gen (R and U only) you need to correct edge orientation and corner permutation. Tricky stuff.

GuRoux said:

i think you have to be very good about saving moves in f2l, quick insert like keyhole should probably be pretty common. looks promissing, how are the PAL algs?

Click to expand...

The one he just posted was 6-gen, which is worrying, but it may just be a bad case.

GuRoux said:

i think you have to be very good about saving moves in f2l, quick insert like keyhole should probably be pretty common. looks promissing, how are the PAL algs?

Click to expand...

xcross is also easier, maybe at least an xcross every single solve? didnt try any solves myself but just an idea.

PAL, as far as the alg in the example solve, looks really tricky to do compared to PLL, and idk if you'll actually save enough time during F2L to make PAL algs worth it.

also i think F2L will require a bit more thinking than before, but nicer finger tricks so i guess it balances, but one of the great things about F2L is how you really dont have to think while you do it.

maybe certain PAL subsets (determined by permutation of the F2L slots) will be bad and others will be okay, so you could influence your slots based on that.

btw i just realized, doesnt inserting into the wrong slots only save like 1 move per slot on average? unless you start getting fancy with keyhole of course, but then theres so much thinking and colour mismatching going on that i dont even want to think about it...

maybe someone can tell me why im wrong here

Smiles said:

btw i just realized, doesnt inserting into the wrong slots only save like 1 move per slot on average? unless you start getting fancy with keyhole of course, but then theres so much thinking and colour mismatching going on that i dont even want to think about it...

Click to expand...

It doesn't save too many moves, but for me personally, it removes many cube rotations. I solve at around 22-23 seconds, and my F2L has an average of 3-4 cube rotations (2 if I'm lucky). There was only 1 cube rotation in my example solve. Plus, this allows a window for mostly <R,U,L> rotationless F2L since you can insert a pair with a bad edge into a separate slot.

The 6-gen PAL alg was probably just a bad case. I generated the E-perm with a couple other F2L Permutation cases.
Adjacent swap: F R U R' U' R U' R2 F' R U' F U2 F' U'
Opposite swap: U2 R L U R' U R L' U R' L' U R U' M' (You can do a R' L here instead of an M')
CW 3-cycle: R2 F R2 U2 R' U' F2 U' F2 U2 R' F' R' U2 R' F U' F' U' (Long and not the best, but it's still decently good (and 3-gen!))

Also, I have some bad news. There are more cases than I thought. Here are all of the subsets of PAL (22 algs per subset).
Solved F2L (one solved case in this subset)
Adjacent Swap
Opposite Swap
CW 3-cycle
CCW 3-cycle
CW 4-cycle
CCW 4-cycle
Double Adjacent Swap
Double Opposite Swap
W-perm

That's 175 algorithms. To be honest, you could've learned full EG + most of OLL by that point. However, people have learned full ZBLL, so learning this is far from impossible.

I might learn the Adj. and Opp. swap cases so I can solve F2L allowing two slots to be misplaced.

Smiles said:

maybe certain PAL subsets (determined by permutation of the F2L slots) will be bad and others will be okay, so you could influence your slots based on that.

Click to expand...

I have found that the two-swap subsets are the best. The opposite swap subset can be done all <R,U,L> 3-gen, and the adjacent swap and 3-cycle subsets can be done all <R,U,F> 3-gen. All of the other subsets would have to be <R,U,L,F> 4-gen.

2180161 said:

Or you could just use Petrus.

Click to expand...

Good old days.

For 4x4

2 opposite centers

1x3x4 block on r and l

create 1x2 center blocks

finish edge pairing while solving centers

3x3.

Any thoughts on this? or is this a terrible idea, or has been come up with already?

2180161 said:

finish edge pairing while solving centers

Click to expand...

lol pls

2180161 said:

For 4x4

2 opposite centers

1x3x4 block on r and l

create 1x2 center blocks

finish edge pairing while solving centers

3x3.

Any thoughts on this? or is this a terrible idea, or has been come up with already?

Click to expand...

Sounds like Hoya in a way.

Michael Womack said:

Sounds like Hoya in a way.

Click to expand...

not really

2180161 said:

Or you could just use Petrus.

Click to expand...

:tu
filler

2180161 said:

For 4x4

2 opposite centers

1x3x4 block on r and l

create 1x2 center blocks

finish edge pairing while solving centers

3x3.

Any thoughts on this? or is this a terrible idea, or has been come up with already?

Click to expand...

The most common method it resembles is meyer. But it more closely resembles stadler or my variant of meyer (shadowslice-meyer).

How do you plan to do the edges with the centres (or edges at all)? The reason I use a 1x3x4 on the right and a 1x3x3 on the left is because otherwise it becomes very difficult to pair edges efficiently intuitively.

Stadler uses algs after centres because doing the edges and centres simultaneously would have a lot of algs when there are perfectly good intuitive(ish) methods that are just as fast/efficient.

Sent from my M1005D using Tapatalk

shadowslice e said:

The most common method it resembles is meyer. But it more closely resembles stadler or my variant of meyer (shadowslice-meyer).

How do you plan to do the edges with the centres (or edges at all)? The reason I use a 1x3x4 on the right and a 1x3x3 on the left is because otherwise it becomes very difficult to pair edges efficiently intuitively.

Stadler uses algs after centres because doing the edges and centres simultaneously would have a lot of algs when there are perfectly good intuitive(ish) methods that are just as fast/efficient.

Sent from my M1005D using Tapatalk

Click to expand...

yeah solving edges and centres would be a pain in the ***. if you go through all the trouble of building half centres, just finish em ***. then edge pairing with only M and U layers would be weird. shadowslice meyer and normal meyer are much better.

Method for 3x3

Solve corners. (2x2 method, or any method you like)

solve F2B

LSE

F2B would be done algorithmically allowing for better look-ahead (setup, then alg, undo setup style)


Example solve: Here


Advantages:
Disadvantages: Now, LET THE HATE BEGIN!

2180161 said:

Method for 3x3

Solve corners. (2x2 method, or any method you like)

solve F2B

LSE

F2B would be done algorithmically allowing for better look-ahead (setup, then alg, undo setup style)


Example solve: Here


Advantages:

Spoiler

Algorithmic approach allows for good look-ahead. reduces alg count from 9 to 8, or 42 to 8 (2-look cmll vs full look) good for a beginner so they can transfer to full roux. (ROUX FTW!)

Disadvantages:

Spoiler

INSANELY HIGH MOVECOUNT,

Now, LET THE HATE BEGIN!

Click to expand...

actually, i think the movecount will be pretty good, this is just corner's first with roux lse, probably better movecount than cfop, let me do some solves.

GuRoux said:

actually, i think the movecount will be pretty good, this is just corner's first with roux lse, probably better movecount than cfop, let me do some solves.

Click to expand...

yeah this doesnt seem all that bad. splits would be like 3-10-2.5, and with a little practice, I could sub 15 this.

2180161 said:

Method for 3x3

Click to expand...

isn't this exactly what CF is?

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?

TDM said:

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?

Click to expand...

Quite a few I think and I don't thing that the would be very nice.

Megaminx alternative LL (reducing algs to learn)

Beginners
1) Orient edges
2) form a 1x2x2 on the LL
3) permute edges (2 algs (or 1+inverse (1/3 chance of a skip)))
4) corners (intuitive)

Intermidiate
1,2&3) same as beginners
4) solve corners (42 algs- adapted 3x3x3 CLL?)


Advanced(ish)

1&2) same as beginners
3&4) combined- 126 cases including mirrors I think (<84 without)

Reasoning: 1x2x2 ridiculously easy to form because the 2 pairs can easily be kept together unbroken so fixing of S2L is easy enough.
2) Less algs (pretty much sells it for me ;-))