I received an email responding to my last post (the one in which I declared this a blog for the mediocre!). I explained at the end of that post that, yes, clichéd as it is, the best advice for learning complicated algorithms is practice, practice, practice. The email asked how I practice. The short story is algorithm-by-algorithm, perm-by-perm. I just learned the Ra PLL, for example: (R U2 R) D (R’ U R) D’ (R’ U’) (R’ U R U R’). It was among the harder perms I’ve learned, and I’m not fast at it — between 3 and 4.5 seconds. So, I just drill over and over and over. Do it, reset, do it again. Here’s a quick video showing it:
Yes, that’s a just-purchased off-brand Stackmat timer connected to CCT on my work PC. “Shanty” form Bright Light Social Hour’s brilliant self-titled album is barely audible in the background.
As I earlier posted, I’ve been learning OLLs and PLLs at a moderate pace, trying to move past the inefficiencies of the Beginner’s Method. About half-way through learning A Perm, a light bulb went off: It corner cycles the same way as the penultimate Beginner’s Method algorithm of R’ F R’ B2 R F’ R’ B2 R2 — with Aa as the far less re-grippy surrogate, and Ab (its mirror) more efficient than serial application of the Beginner’s Method algorithm. (Slowly, it’s all coming together….)
I find the A Perm one of the easier cases to recognize, with its characteristic 2x2x1 block in a top-layer corner, the block having matching colors on either side. I set up by AUF‘ing the 2x2x1 block into the front-left corner, with the 2×2 block matching the middle and bottom layers (as in the diagram to the right). If the headlights (green-xx-green in the below diagrams) are in the back, you do the Aa Perm; if the headlights are to the right, you do the Ab Perm.
Over the past couple months, I’ve tried to exorcise all remnants of the Beginner’s Method. F2L replaced the corner-first/edge-next approach a while ago, and several OLLs have made top-layer orientation a (generally) single-algorithm affair. But my PLLs were still a bit all over the place. I got to the two-look stage pretty easily (with no illusions about being able to one-look it for a while still), but realized that I was still using the Beginner’s Method algorithm for edge cycling — to wit, F2 U(’) R’ L F2 R L’ U(’) F2. Although I got freakishly quick at that algorithm, it required all sorts of awkward hand movements and lent itself to no finger tricks.
Since almost every one of my solves requires an edge cycle — with H and Z perms or the rare PLL skip as the lone exceptions — this bull-in-a-china-shop algorithm had to go. Enter the U perm:
Slowly, I’ve been chipping away at new algorithms and permutations. Last week I learned a faster edge cycle technique (both directions), and I’ll post about that soon. This weekend, I learned more OLLs — the three with the corners correct but edges flipped. I’m not great at memorizing, but these are starting to come fairly easily. More muscle memory than memorization, I suppose. Here’s a quick video with the three:
Over the past few weeks, I’ve posted a series of personal bests — culminating in last week’s 34 second solve. But none of those PBs were evidenced by anything other than a screen capture. So, it was nice just now to record (on my hastily set up iPad) this 39-second solve before diving into my work day.
This was on my re-stickered Dayan Zhanchi, with my custom (homemade) G‑sticker. This nicely showcases my slow but steady evolution — with no remnants of the Beginner’s Method. Advanced cross, F2L, Antisune (OLL 26), and the just-learned Ub Perm for the PLL edge cycle.
adventures in cubing 