Memo Method for 4x4 Centers?

[philkt731]

I'm just beginning to look at doing 4x4 BLD, and I read Mike's tutorial on centers. He suggested when looking at the cycles, for whatever color comes next you choose the next center on that face alphabetically (they start in the top left and go clockwise) that is not yet solved. I was thinking, then, could you not first memorize the solved pieces with the same letter scheme Mike uses (I think this averages around 6). And then for the unsolved centers, instead of memorizing a letter for each specific center, memorize a number 1-6 for which face it is on, most likely in pairs, e.g. 12 52 36 41.... Then, in execution, when you get to a pair of numbers, you recall which were solved and how many times each number has been used before and then you will know which two to cycle with your buffer. Could this work? I hope I explained it well enough

 

[Swordsman Kirby]

It requires you to memorize which ones are solved, which is not necessary in Mike's memorization system. Though in the long run, I guess it's the same.

 

[Ville Seppänen]

Yeah that works, but I don't think it is better than what Mike does. You would have to go through solved pieces every time you recall new pieces, which makes things more complicated, where in Mike's way you would almost instantly know what piece is next. I just find it strange to memorise solved pieces. Blindsolving is for braindeads you know. But sure, try it and see how well it works, maybe it works best for you.

 

[Swordsman Kirby]

Yeah that works, but I don't think it is better than what Mike does. You would have to go through solved pieces every time you recall new pieces, which makes things more complicated, where in Mike's way you would almost instantly know what piece is next. I just find it strange to memorise solved pieces. Blindsolving is for braindeads you know. But sure, try it and see how well it works, maybe it works best for you.

Better than when I used to have my cycles actually go through solved pieces.

 

[blah]

Unless I misunderstood it, I think what you proposed requires a lot of backtracking, right? IMO, that requires brainwork, and that's bad for BLD. The more braindead, the better.

By the way, if you decide to proceed with your memo method, there's this slight thingy I noticed: you gave a random example "12 52 36 41". If you're doing 3-cycle commutators, you might wanna memorize it as "1 25 23 64 1" for obvious reasons, get the idea?

 

[philkt731]

By the way, if you decide to proceed with your memo method, there's this slight thingy I noticed: you gave a random example "12 52 36 41". If you're doing 3-cycle commutators, you might wanna memorize it as "1 25 23 64 1" for obvious reasons, get the idea?

well theres no point in memorizing the buffer at the beginning of the first cycle, so it would start out with two digit numbers lik 12 34 42 53, but once you get back to the buffer and have to start a new cycle, then it would be a three digit number like 345 46 21 or 3 45 46 21 where 3 then is the new buffer. idk thats how i've always memoed 3 cycles

 

[blah]

Yeah I get what you mean, but there's a nonzero probability that the "default" first buffer is already solved, so I just memorize it for safety purposes or for self-reassurance or something like that

 

[Mike Hughey]

I guess I don't have much to add, but I felt like I should chime in anyway. I think everyone else has given pretty good answers. Phil, your method should really work just fine. But I think it's a little less than ideal for getting good times. The benefit is that you have less information to memorize to begin with, since each "entry" in your memo is a number 1 to 6, instead of a letter. (I guess you could say it's the same number of "bytes", but each byte has fewer "bits", so it's less overall information.) But the catch is that you have to think much more when unpacking your memorization. It seems like in general, the best policy is to do as much of your thinking as possible with your eyes open, so you can be as braindead as possible when blindfolded.

I'm really happy with my memorization method because it can often be quite efficient. By not starting with a fixed buffer, but rather starting at the next unsolved piece, sometimes I can get lucky and solve 3 pieces with a single commutator. On some lucky 4x4x4 solves I've had this happen 3 times on a single solve, which is really efficient - 9 centers with 3 commutators!