Faster N Permutations

[mrCage]

Hi

The idea behind J-perm and N-perm are very similar.

Swap 2 c-e pairs twice to achieve the desired effect. In fact a commutator idea at heart ...

J-perm : (d2 = (ED)2 ) - bottom 2 layers together

R' L' d2 L R (first swap)

R' U L' d2 L U' R' (second swap)

(Adjust the U layer, then rewrite in pure UDFBRL notation ...)

N-perm :

1 = R' U L' d2 L U' R (first swap)

2 = L' U R' d2 R U' L (second swap - symmetric to first swap)

(Adjust the U layer, then rewrite in pure UDFBRL notation ...)

Deliberately i didn't do the rewriting/cancellation for these perms (exercise).

- Per

 

[brunson]

R' U L' U2 R U' L

Repeat it twice, do the last L in the first round while you start the first R in the second round. Sometimes when I practice it I can do all turns in the alg as one looong trigger.

2x(L U' R U2 L' U R') makes the mirror.

Those are the alg I use, but I do either a z' or a z (respectively) and invert them so they turn into:

z' (U L' D L2' U' L D')x2
and
z (U' R D' R2 U R' D)x2

I like the way the L/R spins forwards, then back and I can execute the U/D moves with index and ring finger of either hand, as appropriate.

It just feels very fluid and zen. ;-)

 

[Dene]

The problem with using a setup, then another alg, then an anti-setup is that you will never be able to do it faster than the middle alg. This isn't a problem if your targets aren't too high, but if you're aiming for sub2 every time you're making it difficult for yourself.

 

[blgentry]

The problem with using a setup, then another alg, then an anti-setup is that you will never be able to do it faster than the middle alg. This isn't a problem if your targets aren't too high, but if you're aiming for sub2 every time you're making it difficult for yourself.

I'm not advanced enough to be shooting for PLLs that fast. Plus I'm not sure I can sub-2 any alg. My finger tricks and dexterity just aren't that good. I think my absolute fastest PLL is in the 3.3 second range. For me, simply having an N perm that is faster than a Y + cube rotate + U is a big win. Every N I've tried just sucked. So I'm happy to have something that I can remember and can do relatively fluidly.

I get your point of course. Maybe when I get a lot faster, many, many months from now, I'll be down to shaving seconds off of my PLLs.

Brian.

 

[nicoJ]

actually the idea just came of thinking for a fastest method
for example, to do T (in the position of a T rotated 90º clockwise)

if we do R2 U' R2 now we do J1 and we undo the first 3 moves, so we do R2 U R2
this way we solve a T perm (wich is not userfull, because T is faster) when you dont know too many PLLs.
when i was starting, first plls i learned were U, A and J
so with that 3 algs you could solve any PLL with a maximum (and sometimes you just have to do 1 pll) number of 2 plls, wich in any other way all plls except for those 3 (u, a and J) will be of 2 plls to solve it

i dont know if you understand.. i can't explain it better
any questions, ask me


so it came of... "how can i transform a pll to another without doing a pll?"
and N came by itself