Random Cubing Discussion

[Escher]

UM2U2M2U MU2M'U2
UMU2M2U2M'U'M2
UM2U2M2U MU2MU2M2

>_>

 

[Tim Major]

Major Permute ^.^ (Pyraminx crazy method)
Taken from the idea of turning Layer By Layer into Face Permute. I generated all algs for this, as noone else had, but didn't really see much potential over Polish V. So then I thought a preposterous idea of turning Polish V into V Permute (and obviously, this gave me the right to name it after myself ^.^)
I wasn't originally going to post about it, as I planned to learn all algs by Melbourne Summer Open 2011, and then surprising everyone
1. Make non matching V. Estimated 0-3 moves on average (I've had quite a lot of skips)
2. Solve all! Estimated 4-10 moves on average
As rotations are annoying, after first step you have to manually rotate to V on the D face, with the empty slot at DF in these examples.

[18:23] <Nibblr> Pyraminx Scramble #1997: U B L B U' L' B' U B L U R' L l'
B R L (3)
U' L B U' R B' L' R' (8)
(11)

[18:46] <Nibblr> Pyraminx Scramble #1999: L B R B L U R U' R' L' U u r'
L' U' B (3)
R' 2flip (9)
(12)

[18:48] <Nibblr> Pyraminx Scramble #2000: R' U B L R' B U R B U B' U R' B' l u r
U' L' B R' (4)
U' L U L' R U R' (7)
(11)

[18:55] <Nibblr> Pyraminx Scramble #2001: L R U R' B L R U' R U' R' U L' U L U L u' r'
U R' B' (3)
L B' L U' B U L (7)
(10)

[18:58] <Nibblr> Pyraminx Scramble #2002: L R U R B L R L' U R' L' B' u r
R U (2)
U L' U' L U' R U' R' U' (9)
(11)

[19:01] <Nibblr> Pyraminx Scramble #2003: U L R' B U R U' B U' B' R' U L l' r
R L U' (3)
R L B U L' U B' R' (8)
(11)

These had quite bad first steps, but as you can see, this can give 1 look Pyra solves, every solve.

There are 4 distinct sets, plus 2 mirrors.
V.
CW.
Adj swap.
Adj swap with one solved.
Solved.

Then about 30 algs in solved and clockwise, 15-20 in the Adj swap sets.
I know the solved set, and the BR and BL adj switch algs for solved 3rd edge, and flipped 3rd edge with 1 edge solved (I also know a few sexy algs such as L' R' L R L' R L R')
I made my algs with this ( http://www.jaapsch.net/puzzles/javascript/pyraminj.htm ) and tested, changed, added rotations/inverses. For these examples I made a few of the algs then, without testing, so they're prolly not the best.
Now I hope noone makes and learns all the algs before me

Edit: I had all of the Face Permute algs up, but webs.com screwed me over

 

[jms_gears1]

Kirjava should post execution videos of his CMLL algs : D?

 

[Kirjava]

Skipped two steps XD

 

[Anthony]

If it's the right thread, you can't be serious!?!?!

If you are, that's awesome!

Wow. Is this with CFOP, and if so how do you memorize? Do you remember shapes or moves for the F2L pairs (or neither)?

Do you use your teamBLD codes to help yourself memorize?

I'm a bit late to reply because I didn't notice the responses, but anyway..
I actually find the hardest part remembering the cross for some reason (unless F2L is hard) because I do use my TeamBLD codes. As long as there aren't too many cube rotations/AUFs before pairs, TeamBLD codes work great. Admittedly, my 40 was pretty lucky because the F2L was very nice.

 

[Robert-Y]

Mad Kirjava laughter

 

[Rpotts]

lol 2.46 H perm. Too bad, if you expected it could've been like low 7s.

 

[Robert-Y]

T perm: x' (R D R U' R' U) (R2 D' R U' R' U D R D' R') x

Can anyone do this fast?

 

[Robert-Y]

Also, a "nice" L perm I found by hand:

x' (R D' R') F (R D R') F2 (R D' R') F (R D R') x

No U (and u) moves!

My best time with this alg is about 1.5 which is reasonably fast...

(You can see how this was found by hand if you know how it works )

 

[Kirjava]

23:17:21 <+Nibblr> Roux Scramble #266: M' U' M2 U' M' U' M U M U' M U M' U M2 U

THIS JSUT HAPPENED!!!>!

 

[qqwref]

HAHAHAHAHAHA

 

[DavidWoner]

I liek

 

[Kirjava]

Random Human Kociemba stuff;

Awkward:
U2 D R2 D' F2 D2 R2 D' F2 D

Awkward Diag:
D' F2 U' F2 D R2 U' F2 U' F2 D R2 D'

Diag:
U F2 U2 R2 U F2 U2 F2 R2 U F2 U2 R2

Looking for a nice stage 2 solution still

 

[StachuK1992]

Human Kociemba...
Looking for a nice stage 2 solution still

Just a thought: Vandenbergh.
Do E slice first, then treat as sq-1, but oddly?

Sorry, but I haven't really seen your other posts concerning this.
Seems interesting. Time to experiment.

 

[Kirjava]

Just a thought: Vandenbergh.
Do E slice first, then treat as sq-1, but oddly?

It sucks.

I can solve phase 2 in at most 3 algs if I'm going for speed, but anything more than 2 is too many imo.

 

[StachuK1992]

It sucks.

I can solve phase 2 in at most 3 algs if I'm going for speed, but anything more than 2 is too many imo.

Ah, gotcha.
What have been some ideas up until this point?

 

[Kirjava]

I can't remember most of them. Usually **** I think of gets dismissed pretty early.

Some ideas that jump out at me are; Corners -> Edges, 2*OppFace -> Everything, Layer+Corners -> Everything, Ring

I can't think of anything that would work well in most situations. Feels like there should be something though.

Freestyle/Trial and Error is great for FMC, but doesn't work for speedsolving at all.

It's all a shame because phase 1 is so easy and fast

 

[StachuK1992]

Hrm, what struck me first is

RING + DBL + DFL
L6C
EO + DL + DR
L6E

But that's too many steps and kinda not fun.
How fast is step 1 for you?

 

[Kirjava]

like 8 seconds

2*OppFace is pretty interesting. It's like Sq1 Cubeshape. Then the second part would just be a 1 look alg.

 

[StachuK1992]

2*OppFace?
Sorry, don't understand your terminology here. :/