avoiding 4x4 pairity problems

[spy1325]

Is there anything i can do while pairing up my edge pairs, to reduce the probability of a parity error? The Algs for a fliped edge pair is essentially giant, and takes a long time to execute.

 

[Fireman32]

I am wondering the same thing myself.

 

[tenderchkn]

Edge parity is created while solving the centers. I remember Chris Hardwick saying something about odd or even number of turns creates the parity. But in any case, however you try to recognize it, it'll take you longer than the 5 seconds or whatever it is that it takes you to do the parity alg.

 

[Fred]

You could count how many edges are oriented correctly according to the common BLD definition as you pair them up. Or you could use a different method.

 

[Erik]

The parity is already decided after placing the first 3 centers...

 

[Fred]

Oh, that makes sense. Once you have three centers, you can't do a slice move to change parity. That's why I leave a slice free until the very end.

I'm still wondering, though. Is it advantageous to know at the end of the edge-pairing stage that you have a parity problem? I know it would be a pain to count how many edges are oriented correctly, but I think that with practice, it wouldn't slow you down too much.

 

[pjk]

I'd be interested in learning how you "look ahead" to prevent a parity.. what to look for, etc. Eventhough it may take longer than the time to fix, I am still interested in learning. Thanks

 

[Fred]

I don't look ahead to prevent parity. I use a different method from reduction, which solves a slice last, so my parity fix is just a slice move.

For the reduction method, I was suggesting the idea that you count the number of "correcty oriented" edges as you pair them up, and maybe that could help in some way. It seems to me that it would be an easier alg to do a parity fix at the beginning of the 3x3x3 stage than near the end.

 

[Erik]

so why do we all still do the full parity algorithm after pairing up on 5x5? Apperently there is not a faster alg (yet)

 

[Joël]

Originally posted by Erik@May 2 2007, 08:50 AM
The parity is already decided after placing the first 3 centers...

I am afraid that is incorrect, Erik. Parity is 'decided' (more like fixed) after solving all the centers.

 

[AvGalen]

I always thought parity couldn't be avoided. I read that parity could occur, even on a 4x4x4 super-cube. That would mean that centers have no influence at all on the 4x4x4.

(I read parity can be "avoided" by using a centers-last method. This way parity-fix could be as simple as 1 move, but still not avoided)