Nice! It's such a fun method. It seems like it could be actually be a half good speedsolving method, or maybe that just me disliking the CFOP disease.
What method do you propose I improve on? I am trying to get as many methods sub20 average of 12 as I can. I have 6 methods so far being CFOP, Petrus, Columns, Roux, nOOb, and Belt. I'm thinking of doing Waterman.
Yes I do Colums First. I do pairs first(usually in two look) then I fix centres, then do OCLL, solve and orient edges at the same time, then PLL. If I know the COLL I do it.
PB is an 11.65 and 14.xx(nl), average of 5 17.xx and average of 12 is 18.56.
Like I solve the first block as a whole block then solve the opposite centre. Really, the only benefit to me is that I can plan much more of the first block in inspection if I forget about solving it centres first.
Yeah, I'm still active on FFR. It's mostly on the forums though; I don't play all that much these days (mostly because of a spinoff FFR site I've been working on, indeedengine.com, where we have a whole bunch of our own files for people to play).
Oh well, too bad it doesn't work. Normally it has a <body onkeydown="function"> thing and Firefox 9 has a really bad bug that breaks those, so I had to add one in the code to make it work. It should treat that as just pasting over the data it already has, so it only triggers once, and I was thinking that maybe in your case they were each triggering separately (so each key press would do the move twice), but I guess that wasn't the case since I removed the one in the body (at least in qcube.html) and it didn't seem to have any effect. I'm really not sure what else it could possibly be.
I looked up "exception to the rule." on like..5? sites or so wondering if what you wrote was positive or negative. Thanks for putting me back on the right track tho :) Could have saved myself the hassle ;)...Damn language barrier...
Efficiently, no; for me qcube-gray is all about minimizing the trial and error you have to do. For instance, I generally prefer to solve one piece at a time (well, more in the case of centers) directly. It's often useful to only know one sticker of something; for instance if the first layer of 3x3 is solved, and an edge matches F but not U, you know you can insert it in FL or FR. The same idea applies to stuff like "F4L" stuff in K4 etc. And for the first layer, you can use F/F2/F' moves to test if something belongs there or not, before actually trying to insert it.