--- In [email protected], "thomkirjava"
<[EMAIL PROTECTED]> wrote:
> I solve M differently as I optimised my solving M for each
> alg I use (yes, I actually went through every orientation and worked
> out different ways of doing finger tricks and algs. I can solve the
> two opposite edges flipped on top in 1.05 :)
Eventually, I would like to do this for all of my algs, especially for
the L5E. There are only 16 or so, and nearly half of those are just
mirrors, so it shouldn't be difficult.
As for the two opposite edges, eventually, the method I'm developing
will avoid those cases that aren't oriented or the 3 edges on top 100%
of the time.
> Mostly alternative ways of going about something. I especially wanna
> see your LSE approach.
> Alternative algs! CMLL mostly with COLL recognition helpers :)
> Oh, and have the backwards verssions to setup the cases for easyer
> practise.
> Different stratergies (like turning it into a faux-CF and things)
> Mostly help with the F2B :S little tricks like to swap two edges or
> flip two - they're always fun.
I will list all of the algs I use, and provide more than one alg for
each case most of the time. My method involves COLL, so I will
obviously post those algs. I can list inverses for setup, too.
I got a solve yesterday where the D corners were all finished before
the solve, so I used COLL to fix the others. I finished with a
generic corners first solution (not very efficient and averages ~65
moves) for a 23.36. I think I may include a page about tricks like
this when I finish.
> Not so much fewer moves, but faster sequences than I'm using now. I
> solve in about 57 moves, so I could drop 9 somewhere I'd like to find
> where I could get rid of them aswell as speeding up though.
We got the (M'UM'U')*3 alg to over 12 moves/second, and I'm sure we
can optimize the L5E algs similarly. These will be lightning fast
once mastered.
> As many as it takes to be faster :) The only limit is time really, not
> the amount of algs. Although 100 sounds a nice value to be going on
> with. (long algs, not things like M'UM - they don't count as hard to
> learn :P).
The full system would use about 78 distinct cases, and about 60
mirrors. 138 total is not so bad, especially when 40 of those are
COLL. Almost all of the rest are algs to insert a pair and orient
edges, so they are much shorter than LL algs.
> LSE would be hard to do with face diagrams. I think java is your best
> bet, but usually it crashes my browser to have more than one running
> on a page. I think it's best to do it similar to gilles (imagecube?)
> and show the alg, I mean you can work out what it does just from the
> alg, really.
The COLL algs will wind up with a diagram as on Lars Vandenbergh's
page. I think I will use imagecube images for almost every other case.
> Not really. It'd be nice to have a section on helpful tips and tricks.
> Such as, inverted and opposite block building. I like how I'm now able
> to start on one of four blocks, and then I'm even able to build the
> second upside-down. It's gotten to the point where they just all feel
> like the same block. Doing the corners after that is also easy, and I
> can always solve it in two or less moves, recognition time is almost
> zero, I have a handy little system I use to check.
I would be interested in learning about how you recognize mixed block
corners quickly, as this has been difficult for me. That is
definitely something I want to go on my site. Please email me so we
can talk more about this. [mikebennett_one] AT [hotmail.com].
> ~Thom
-Mike
P.S. I'm still interested to hear what Jason and Gilles think.
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/
