From:

~~~ ~~~ Subject:

I've hardly ever found two people who solve the same way. The most

common method I've found around Stanford is to go for all the edges,

then the corners. My own method is to get the top corners, then the

top edges (via operations like -l r f^2 l -r & similar tweaks) then three of

the middle edges (via things like l^2 u^2 r^x u^2 l^2, where x=1, 2, or -1),

then get the other four corners in the right place (requiring at most one

corner-swap, for which I have a fairly simple macro), then get those four

corners oriented using the "rFUfuR" commutator macro, which also randomly

alters the remaining middle edge, then get the last five edges, then flip

edges as necessary. My typical time is about 5 minutes. "Best" times are

meaningless, since anybody can luck out once or twice; the best measure

of your solving speed (in my opinion) is your WORST solving time over

your most recent ten or so attempts. Of course, for real fun, pick some

pretty pattern and solve to it without going through the normal solved state!

-- Don.