Non-Permuting OLL Algorithms, Improving PLL look ahead

[Ranzha]

Instead of this ghastly beast of a substep, when I get F2L done, for a good number of cases, I recognize not only the OLL, but the CP. By doing so, I can predict what the CP will be after I perform the OLL alg I use for that case, given I know how it affects corner permutation.

 

[sa11297]

i do this

find the oll as fast as i can - execute oll as fast as i can - find pll as fast as i can - execute pll as fast as i can

seems like the best way to me. i avg 5-6 seconds for the last layer and it would be faster if my tps didn't suck

i am the same way. i just do everything normal and it takes 5-6 seconds.

 

[Talon2461]

My opinion probably doesn't seem like much, but it seems that taking a lot of time to learn how to recognize PLL during the OLL seems like time you could spend on trying to do other things to improve on your time. I would say that just doing it normally produces the fastest results for (probably) most cubers.

 

[jiggy]

http://www.speedsolving.com/forum/showthread.php?22862-Random-Cubing-Discussion/page8

But this way, you do PLL, ignoring orientation, then Pure OLL.

Ah, good find!

Yeah, generated the algs and didn't post them (except some of the interesting short ones) because they suck.

PureOLL is not a good step.

Fair, but I still thought it was an interesting idea. Ideas don't always have to have a practical application...that would be boring!

 

[MaeLSTRoM]

The 2 and 3 G algorithms have high movecount. Other than that its only about 3-4 moves extra than the original algorithms(Not using double turns/slice turns).

For example if you made the normal headlights OLL 2G <R,U> then it would be this optimally

U' R' U R' U' R' U R U R' U2 R U2 R U' R (16f*)

Woah, wait a minute. For headlights an Optimal Pure corner rotation alg is:
L' U2 L U L' U L R U2 R' U' R U' R (14f*)
Its less moves and more finger trick friendly than yours and <RUL> can still be very fast.

 

[irontwig]

Instead of this ghastly beast of a substep, when I get F2L done, for a good number of cases, I recognize not only the OLL, but the CP. By doing so, I can predict what the CP will be after I perform the OLL alg I use for that case, given I know how it affects corner permutation.

+1 And this is really easy because many OLLs solve the same CLL case.

 

[JLarsen]

+1 And this is really easy because many OLLs solve the same CLL case.

+1

Knowing how each OLL affects EVERY piece may not be worth it but CP? Definitely. Given your times are even fast enough to start worrying about optimizing your last layer that is.

 

[jms_gears1]

+1

Knowing how each OLL affects EVERY piece may not be worth it but CP? Definitely. Given your times are even fast enough to start worrying about optimizing your last layer that is.

+1

And you dont even really need to know how they effect the CP. You only need to learn to recognize one case, because theres only one CP for every OLL alg that will lead to a corners opp swap

 

[StachuK1992]

Just make your OLL algs all RUr and you're good to always know CP.