Best Method for OH

[Morley]

Hello everyone, I finally decided to join after visiting here for about a month or so. Recently I have given up 2-hand completly for oh until I have a faster pb and average for oh.

I learned a LBL method when i first started cubing in July 2007. Since then I have learned Fridrich F2L and Some OLL and PLL algs. After that, mabey a month ago, I switched my main method over to Petrus F2L. Once i started OH I have been have some problems with Petrus. Mostly with fixing bad edges, because of the F moves. I was just wondering if anyone had any suggestions for finger tricks on fixing the bad edges. My other option is to learn VHF2L, because I want to eventually learn COLL.

Thanks for any help or suggestions, I look forward to posting here some more. It looks like a nice community.

Morley

 

[cdzoan]

no idea, I don't use petrus

 

[Pedro]

why not make it Petrus style until the 2x2x3 and complete the cross and F2L?

 

[alexc]

why not make it Petrus style until the 2x2x3 and complete the cross and F2L?

That's not a bad idea.

 

[Lofty]

I think the best OH method is the same as the best 2H method.

 

[ExoCorsair]

Rowe Hessler builds a 2x2x2 block and then decides on the spot to go X-cross or Petrus, iirc...

 

[Johannes91]

why not make it Petrus style until the 2x2x3 and complete the cross and F2L?

Why *would* anyone do that? It's not Petrus at all and loses many of the advantages that Petrus has (like finishing the F2L using just RU and easy LL). That would be just Fridrich with a double extended cross.

If you want cross and ce-pairs, then use Fridrich F2L. If you want block building, use Petrus (or Heise or ...).

I have been have some problems with Petrus. Mostly with fixing bad edges, because of the F moves. I was just wondering if anyone had any suggestions for finger tricks on fixing the bad edges.

In that step, you should hold the cube so that the 2x2x3 block is at DB and the two free sides are U and F. So you can often use 3-movers like R U' R', L' U2 L, etc. to fix the edges.

If you don't like F moves, you can avoid them by using double-layer turns. For example, l' U2 l instead of R' F2 R and r U' r' instead of L F' L'.

 

[Morley]

Wow thanks Johannes I never realized you could fix the edges like that, it never occured to me. I think ill try the double layer turns for a while, even though I will probably just drop the cube.

 

[Lofty]

Ok I have been thinking about this some with out finding too much. If there is actually a method better suited for OH.
A few ideas:
Orienting edges first has been a suggestion. If this is done the entire solve afterwards can be done with LUR and no cube rotations. The cons of this are moves are added to the start of the solve to orient edges and moves have to added to f2l to preserve them. From my COLL optimized for OH I learned LUR is not always best.
If edges are oriented (not necessarily permuted) and corners are permuted (not necessarily oriented) then the result is a 2-gen ZBLL alg which I have not found a non-2gen alg to be faster than a 2-gen one. I don't know what it would take to make sure your corners were permuted but if edges are already oriented at the start it shouldn't take too much.
One problem is things like no cube rotations and only LUR cannot be generalized to always be faster. Being obsessed with OH I wouldn't mind making and learning a method made solely for OH, anyone actually good at making methods or has ideas want to help?

 

[Dcuber]

u might not want to "give up 2h" completely
do occasional ones, OH tires ur hands out
petrus is very good if u get below 50 moves
I think COLL is worth learning

 

[Lucas Garron]

I actually happen to think that MGLS with pre-orient (F2L in URL) is the best OH method, but only if you're flexible in use (like, don't use it if EO is weird or there is a good non-orienting cross/xcross).
I might be a bit biased, though.

Trying one: 24.99

 

[Lofty]

haha OH doesn't tire my hand out
So what MGCLS? I'm more a fan of my COLL then a 2-gen edge cycle I need to get back on learning them... I forgot them all... :-(
What would it take to permute corners disregarding orientation?
I'm going to regenerate my COLL spending more time on it. Trying to add double layer turns, cube rotations, knowing not all F, B and D are bad, etc.

 

[Johannes91]

Orienting edges first seems very promising to me, too.

 

[Swordsman Kirby]

What is the best 2H method...?

 

[Kenneth]

What would it take to permute corners disregarding orientation?

If you want to preserve edges?

Niklas, 7 turns makes a "A PLL", "N PLL" is harder, I think 10 or 11 optimally.

 

[Swordsman Kirby]

Not preserving edges, I like the following algorithm:

U

Alternatively, we can use:

U', U2

In HTM they are equal in length!

 

[Dcuber]

Orienting edges first seems very promising to me, too.

yea, if u don't know full oll, try doing that during f2l

 

[Lofty]

Orienting edges first seems very promising to me, too.

You just said something nice about something that isn't Petrus! lol.
And I don't quite know what a U adjustment will do to permute all the corners...
I know someone was talking before about Orienting at the start of the solve. does anyone know that stats for what it takes to orient at the start of the solve?

 

[Kenneth]

One way is to do a 2x2x3 first, then to permute corners first place RFD and RBD, use FRUR'U'F' and Niklas to permute LL corners, use M' U/U' M and M F/F' M' to orient edges in U and F face (if the 2x2x3 is at BD). After that the rest is possible to solve 2-gen.

Only problem is the M turns used to orient edges but the algs are short (use the thumb to do M in one turn instead of go R' r). The two algs orients four edges at the time, the first 1 edge at F and 3 at U, the second 3 F and 1 U, if it is another case you will have to repeat and do U or F setups between.