OHLL (One handed last layer)

[Faz]

OK this is just an idea, and I'm trying to work out the pros and cons of this idea.
Please note, this is not intended for slower solvers, mainly for those of us that are sub 20 seconds for OH.

- OH PLL's are generally very annoying to solve.
- Except the edge PLL's
- Basically, my idea is to solve OLL and CP at the same time. This is 6 cases per OLL. Some of them, however, you already know.
- 55 OLL's (Yes, they all have different COLL though )
- 42 COLL cases (You should know all these by now, unlike me)
- Other random tricks you may have picked up

Now, this makes 6 cases for each OLL at the maximum. Some may be reflected, but I don't think that reflecting algs would be a good idea in OH.

So, 330 cases - ~100 - minus other tricks = Approximately 200 algs (Yes, I am being optimistic here)

Also, if I were to learn them, I wouldn't learn them for 0 edges OLL's, as I rarely, if ever, get 0 edges cases in solves anymore. That takes away ~30 algs.

These algs can not only be used for OH, but would also work really well on 4x4, and big cubes, where a bad G perm can cost you a World record

A few setbacks

- I haven't tried generating any algs yet, and I'm not sure how long they will be. When/if I get algs, I will make them as fingertrick friendly as possible.
- Full edge control -> COLL -> EPLL might work just as effectively.
- Recognition might not be easy, but it would be easier than NeoPLL


Also, if this goes ahead, I'll call it FazLL

Any thoughts?

 

[Innocence]

OK this is just an idea, and I'm trying to work out the pros and cons of this idea.
Please note, this is not intended for slower solvers, mainly for those of us that are sub 20 seconds for OH.

- OH PLL's are generally very annoying to solve.
- Except the edge PLL's
- Basically, my idea is to solve OLL and CP at the same time. This is 6 cases per OLL. Some of them, however, you already know.
- 55 OLL's (Yes, they all have different COLL though )
- 42 COLL cases (You should know all these by now, unlike me)
- Other random tricks you may have picked up

Now, this makes 6 cases for each OLL at the maximum. Some may be reflected, but I don't think that reflecting algs would be a good idea in OH.

So, 330 cases - ~100 - minus other tricks = Approximately 200 algs (Yes, I am being optimistic here)

Also, if I were to learn them, I wouldn't learn them for 0 edges OLL's, as I rarely, if ever, get 0 edges cases in solves anymore. That takes away ~30 algs.

These algs can not only be used for OH, but would also work really well on 4x4, and big cubes, where a bad G perm can cost you a World record

A few setbacks

- I haven't tried generating any algs yet, and I'm not sure how long they will be. When/if I get algs, I will make them as fingertrick friendly as possible.
- Full edge control -> COLL -> EPLL might work just as effectively.
- Recognition might not be easy, but it would be easier than NeoPLL


Also, if this goes ahead, I'll call it FazLL

Any thoughts?

Or you could just use Petrus/ZZ. Or ZBF2L. Or even VHF2L.

I don't know though, if you think it's worth it, go ahead. Probably IS worth it at sub-16 level I guess.

 

[Diniz]

Yeah, i dont know too if worth it. Iam not fast at all (~25 avg OH) but i have more difficulty on OLL then PLL.

But Innocence, i dont think you got the spirit of the thing, he doesnt wants better OLLs (edges oriented case) he wants better PLLs...

 

[amostay2004]

Yeah, i dont know too if worth it. Iam not fast at all (~25 avg OH) but i have more difficulty on OLL then PLL.

But Innocence, i dont think you got the spirit of the thing, he doesnt wants better OLLs (edges oriented case) he wants better PLLs...

He meant that by getting edges oriented OLLs, you can just do COLL and avoid the need to learn what faz is suggesting.

 

[plechoss]

It's a good idea, I use this pretty often but I think that generated algs wouldn't be as good as normal OLLs
I would go for EO + CP -> 2GLL but 2GLL algs are sometimes slower than the worst PLL you can get
And I usually try to go for Full edge control -> COLL -> EPLL

 

[Erik]

-yawn- nothing new faz, Im using this for over a year now already (ask Rowe and some others whom I've told) but I dont know that many algs yet. Its partially why I get so many PLL skips in FMC

Basically I came to the conclusion that its a good idea to learn only the fast algs and at least the ones that normally would lead to a diagonal perm.

For instance where the normal alg for OLL would be: F URU'R'URU'R'U F' the alg that solves the diagonal corners would be: (U) R'U'R'FRF'RU'R'U2R, a very fast alg that improves your LL.
However, for other cases it is just faster to do OLL PLL since the 'special alg' to solve the OLL plus corners is just too slow.
I'd say if you learn them all you'd not much faster because of those bad algs.

As for the name, I like to refer to it as COLL myself since this is REAL corners+OLL and the thing that we now call COLL is better named CCOLL since it handles orientation of Corners * Solving the corner OLL (or sth like that w/e).

 

[whauk]

some calculating:
i have an average of 2.53 for OH-PLL (using best of 3 and chance of occurence). using this method it would be 1.28 (!). still Hperm is very slow (2.8)
i think i will not learn them all. only the fast ones (like eriks). but where can you find them?

 

[Escher]

I've been using pretty much the same thing Erik has said for quite a while, I think I made a thread on it and everybody flamed me for it. Oh well.

There are a couple of OLL cases (apart from COLL) that I know all the corner cases for, most of the algs are good, since they are mostly just inverses of other OLLs...

 

[Jai]

Kenneth has been using this for around 2 years. http://www.speedsolving.com/forum/showpost.php?p=29981&postcount=2

 

[DavidWoner]

Yeah I do this enough to prevent diagonal corner swap cases, since J, G, and T are still pretty fast for OH. I need to learn some more cases though.

However, for other cases it is just faster to do OLL PLL since the 'special alg' to solve the OLL plus corners is just too slow.

Pretty much exactly this.

some calculating:
i have an average of 2.53 for OH-PLL (using best of 3 and chance of occurence). using this method it would be 1.28 (!). still Hperm is very slow (2.8)
i think i will not learn them all. only the fast ones (like eriks). but where can you find them?

But your OLL would be way slower. Just try the algs on the wiki to see what they do to corners.

 

[Faz]

Thanks for the information guys! I think i'll probably just learn a few algs off the wiki, as DavidWoner suggested.

Also, I've tried using Acube, Cube Explorer, and Ron's solver to generate these algs, but they don't seem to like the grey edges. Any other place I might be able to generate a few for special cases?

 

[miniGOINGS]

Thanks for the information guys! I think i'll probably just learn a few algs off the wiki, as DavidWoner suggested.

Also, I've tried using Acube, Cube Explorer, and Ron's solver to generate these algs, but they don't seem to like the grey edges. Any other place I might be able to generate a few for special cases?

http://www.speedcubing.com/CubeSolver/CubeSolver.html

I put that [bleep] on everything.

 

[Faz]

That is Ron's solver.

 

[miniGOINGS]

That is Ron's solver.

I know. I just don't understand how it doesn't like the greyed out edges.

 

[Faz]

Well, firstly, it just flickers through the algs for some reason. And when I prtscr, and copy it into paint, the algs just mess up my cube.

 

[miniGOINGS]

Well, firstly, it just flickers through the algs for some reason. And when I prtscr, and copy it into paint, the algs just mess up my cube.

I've never had those problems, do you want me to find algs for you?

 

[Faz]

 

[masterofthebass]

These algs can not only be used for OH, but would also work really well on 4x4, and big cubes, where a bad G perm can cost you a World record

ya cause your OLL will never be bad. ever.

 

[miniGOINGS]

Which moves to you want to restrict it to? Or should I run all of them except slice?

 

[mr. giggums]

Make the upper edge on the green side yellow. It shows that because it is an impossible case.