Random Cubing Discussion

[qqwref]

Haha, you're saying a 9-move algorithm "can't be that bad" if you can consistently sub-2 it? I think you are WAY too fast for that to make any sense

EDIT: Here:
x r' R2 D' R U2 R' D r R2

 

[Kirjava]

Ooms method o_O.

 

[theace]

what's a good square 1 method? The one i use gives me an avg of 1:30

 

[qqwref]

http://www.cubezone.be/square1.html

...but for the edge permutation step, don't learn everything yet, just start with any one parity and the adjacent/adjacent, opposite/opposite, U-perm, Z-perm, and H-perm algorithms.

 

[amostay2004]

A fun and fast 3 cycle alg

(U D' R) U' M' U2 M U' (R' D U') - my fastest is 1.15 with it.

Also (U D' R) U M' U2 M U (R' D U') for the other way round. My best is 1.31 with this.

 

[jms_gears1]

OMG BEST 2X2 YPERM EVAZ

F(RUR'U')(RUR'U')(RUR'U')F' (R2U2RU2R2)

oh oh oh and A GOOD HPERM:

FB' F(RUR'U')(RUR'U')(RUR'U')F' (R2U2RU2R2) F(RUR'U')(RUR'U')(RUR'U')F' (R2U2RU2R2)F'B

 

[qqwref]

Edge 3-cycle + two corner flips.
(R U' R U' R) D (R' U R' U R') D'.

Similar, edge 3-cycle + four corner flips.
(R U' R U' R) D2 (R' U R' U R') D2.

 

[Kirjava]

holy balls they're not bad for CMLL

 

[Kirjava]

Ok, anyone considered a cubeshape+parity method?

(Of course you have - you've just never bothered to see if it works)

You'd have to like, find a way of tracing/detecting in each shape and solving the shape differently depending on the parity.

Any ideas?

 

[qqwref]

Memorize Mike Hughey's translation matrices, then track the cycles of 8 corners and 8 edges, in inspection, to determine if the cycles will have parity or not? Sounds incredibly impractical.

 

[Kirjava]

Indeed. So maybe there is a way to trace without using the translation matrix for the shape? Trace in the random shape, not the cubeshape.

 

[blade740]

There's absolutely no way to recognize it in 15 seconds.

 

[riffz]

There's absolutely no way to recognize it in 15 seconds.

Prove it.

 

[qqwref]

You're multiplying two permutations on 16 elements and then checking to see if it's odd or not. Not gonna happen unless someone spends literally years working on it. I don't think you can check each one separately, either, because you'll have different numbers of pieces in each layer - you have to trace through the matrix.

 

[blade740]

There's absolutely no way to recognize it in 15 seconds.

Prove it.

Try to tell the difference between this and this. You have to literally look at every single sticker on every single piece and determine what the overall permutation will be after you solve cubeshape.

 

[riffz]

There's absolutely no way to recognize it in 15 seconds.

Prove it.

Try to tell the difference between this and this. You have to literally look at every single sticker on every single piece and determine what the overall permutation will be after you solve cubeshape.

That's not a proof.

(I was just picking on you for using the word "absolutely")

 

[theace]

Is the COLL, CML and all the fancy new stuff better than pure CFOP?

 

[jms_gears1]

Is the COLL, CML and all the fancy new stuff better than pure CFOP?

Whaa....
Well knowing more algs cant hurt...


@Kirjava, could you make videos of your CMLL algs soon? It would be awesome sauce.

 

[Kirjava]

Uhh, maybe.

Why do you want the video though? You already have the algs.

 

[jms_gears1]

Uhh, maybe.

Why do you want the video though? You already have the algs.

Because I want to see how you execute them. Some of mine are ok but im realllly slow at them.