Question about 3OP method

[crazyhippo]

I've been learning from Macky's tutorial, here:

http://www.cubefreak.net/BLD/3OP_guide.html

and I want to ask...

When you're permuting the corners or edges, what if your cycle is longer than a 3-cycle?

And I don't understand that Cycle Reduction Rule. Does it basically mean that you take out the 2nd and 3rd number in the cycle, if your cycle has 3 or more numbers? If so, how does that work? Doesn't that sort of screw up the cycle...?

Or am I just being n00b...

 

[Ellis]

If your cycle length is 3, then the cycle will be complete. If it's longer than 3 then you're not really solving 3 pieces, only 2, and setting up the third piece to continue in the rest of the cycle. Taking out the 2nd and 3rd number after you complete a 3-cycle is because they are already solved. So if you have a cycle 1-2-3-4-5, you cycle 1-2-3, then 3 is in 1 position ready to be cycled as 1-4-5.

I haven't actually learned 3OP, but I don't think the description is inaccurate.

 

[crazyhippo]

Ok, now that makes sense. Thanks a lot!

 

[crazyhippo]

Oh, and another question.

Also about the permutation, how do you figure out the set up moves? I can't do it at all!!

 

[Ellis]

Follow the move restriction that macky gives. For corners, he says set up moves must be (UDF2B2R2L2), which means you can do any move besides a quarter turn on the F B R and L faces to setup the corners.

 

[boiiwonder]

You just have to figure that out by using the restrictions and figure them out .

 

[rjohnson_8ball]

Macky gives some examples for setup moves. One useful algorithm his pages taught me was the Caltech maneuver (R' F R F')*3 which swaps the UFL with UBR and UFR with UDR while keeping the corner orientations correct. That is sometimes easier to set up for than an E perm or H perm (plus U2).

There are 4 stages to 3OP. After you get confident with each one separately, try to do 2 stages at a time, then eventually all 4 stages with BLD.

Ellis explained very well how to work with long cycles. Remember, as the cycles get reduced, you might end up with pairs. When you have 2 pairs you can set up for a PLL.

When it seems too complicated to set up for a PLL, you can do what I do -- convert one pair into another to make things simpler. For example, you might want to set up for a T-perm if that PLL is easiest for you to swap 2 edges and 2 corners during BLD. If corners UBL and DFR need swapping and your T-perm expects UBL and UFL to be swapped, then do the cycle (UBL, DFR, UFL). That will solve cubie DFR and cause UBL and UFL to need swapping. Likewise, edge pairs can be converted to easier pairs.