Centers with comms are much easier - it's obvious. Have you read Ollie's thread about it? There are only few cases for comms. Others are setups + the case you know.
Yes I've seen that thread but didn't work through the examples, and haven't really got my head around comms yet. I followed Noah's tutorial. U2 seems a good beginners' method because it follows the same principles as r2/OP. It didn't take long to pick it up.
Grzegorz Jałocha told me - and he says the same in his polish tutorial - the he did centers always with the same rule: starting from the same point on each side and adding pieces clockwise. If the piece is correct he was starting from another etc.
Noah says the same, and that's what I tried at first, but I find it harder to remember where I am on a clockwise sequence than I do to think through a tiling strategy (although maybe it's slower). For example, let's take the R face, and suppose it has centres R, B, U, L (face letters, not sticker letters). Now on my B face let's suppose there are two U pieces. I have an R piece in the buffer to solve, but which target to shoot to? Well, i would replace the U piece last, because I want to avoid a cycle break. I would also defer replacing the B piece, because that will speed up the dislodging of the U pieces from the B face. So I would fill the R positions in order L, B, U.
Of course D and U centers are easier with comms but for me, when I was doing it with U2, D centers were pain in ass. U centers were much easier - it was almost the same alg for Ubl and Ufr. But it's my personal preference.
Noah's D alg is longish but the same for all four pieces, with setup, so that's easy enough. The A/C alg seems just as bad to me, but I have been wondering, why not just rotate the U face to move a different sticker into the buffer rather than doing a swap, and track the cumulative rotations? Is that what is meant by a "floating buffer"?
My only advice is: practice. After some time your memo will shorten.
Well yeah, my success today was 18min memo and 9min execution. On 3BLD I'm still spending 1.5-2min on memo. So to get faster, that is clearly my main weakness to target. Once my memo is faster than my execution, then I'll start optimizing execution. For 4BLD, I only really care about accuracy right now. For 3BLD I'd like to get faster
.
About corners with OP in 4bld. I like the idea doing corners in the beginning of the solve. But the problem is that y-perm is not center safe and rotates centers. I was always trying to make corners with the number y-perms divisible by 4. When you know few comms it is easier to do it. For example if there are 6 targets I do 4 with y-perm and 2 with comm. Now I do them all with comms and I always start solving corners first.
Why do corners first? I'm the opposite - I like doing corners last because it feels like the final straight, flat mile on a marathon
.
