Hey Jon, There absolutely is a way to avoid the parity error, but you're not going to like it. The only ways I have found to avoid the edge parity 100% of the time are to 1) count the pseudo-edge cycles during inspection or 2) count the number of inner face quarter turns in your scrambling alg (which is cheating btw)
As for counting pseudo-cycles basically during inspection you count the cycles to pair up a wing edge with it's partner wing edge, not how to pair them up correctly. This is exactly how you do it for the 4x4x4, and since the the 5x5 contains a 4x4 it works there too. This is basically like following the cycles of a 3x3x3 cube, so max you have to follow 12 cycles, but it's very hard to locate the full 12 "pieces" and I have found that the fastest I have ever successfully counted pseudo-cycles is a little over 20 seconds. I think with a few months training that maybe this could be done in under 15 seconds most of the time, and maybe with mastery in under 15 seconds. The problem is, it would require most of the 15 second inspection, and you wouldn't be able to plan any of the first steps of your solve, since you would have been too busy counting pseudo-cycles. Also, I doubt the accuracy of this method in a competition setting where nerves become a factor and you might miss a piece, or miscount the cycle. Also, yeah your method is populary known as "centers-first", and is a good method. The alg Craig posted is good to fix parity. There are a few other ones for the 5x5, but I don't know them very well. There is one that starts with your (r' U2)x4 r' and then moves edges back though. Check bigcubes.com for other cool algs. If you're crazy enough to try counting pseudo-cycles I'll explain in depth how to do it, but from my experience it's not worth it. Happy cubing, Chris --- In [email protected], "harveyjonathon" <[EMAIL PROTECTED]> wrote: > > My second post, and inside 10 minutes... > > I recently got a 5x5 cube, and at the moment, can solve it > everytime, but I feel my method isn't 'tight' and elegant enough. > Basically, and as is pretty inuitive, I complete the centers, build > the edge triplets and solve as a 3x3, which requires minimal thought > (and practically 0 extra algs). The 'center solution' is naturally > slow, so I'm going talk exclusively about associating the edge > triplets. I'd like to hear from anyone who has a relatively > EFFICIENT way of avoiding the 'parity' error (or, more correctly, > when you must swap TWO small edge cubies, instead of three). To > correct this, I simply perform RrU2 five times, which muddles up a > few edges, but yields a cycle of three. This, I find, is incredibly > inelegant. Are there any ways to COMPLETELY avoid this possibility. > I initially thought looking ahead when I had four edges to go > (to 'combine') but this does not seem to have any effect. > > So what do you think?? > > Jon > Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/speedsolvingrubikscube/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
