Every once in a while, my hobby intersects with my profession. That was the case a year ago when I wrote about the controversy between Dayan and Seven Towns over the latter’s (largely exaggerated, I believe) position that it had a copyright/trade dress claim against any cube maker using the standard yellow-white/red-orange/blue-green color scheme. And it’s the case again now, with my recent project to decorate my new office. I figured that the lawyer who deals with patents by day and speedcubes in his free time might as well have some topical art.
I downloaded a variety of cube related patents, ranging from the very first related patent of which I’m aware (from the 60s), the Japanese magnetic 2×2 patent from the 70s, the Hungarian and US Rubik patents from the early 80s, the (in)famous Verdes V-Cube patents, and the (relatively) recent Dayan ones. In Photoshop, I assembled them into a composite with the abstract pages and key diagrams, had it printed at 24×36 by Kinkos (for $4.50), and framed it (via Aaron Brothers for $35). Voila! Legal-cube-nerd art:
Click on the thumbnail to expand. I’m not convinced there’s a high demand for this sort of thing, but, just in case, here are links for downloading it:
It’s been a little while since I felt like I’ve accomplished much in terms of cubing, especially after plateauing at about 30-35 seconds on 3×3 solves. So, I decided to revisit PLL time attacks, and, after a few days of practice, got a good outcome: an on-camera 72-second PLL attack that started and ended with a solved cube:
By way of brief background, PLL is the last step in CFOP — once you have a solved bottom face, solved bottom two layers, and solved top face. All that’s left is to get those four edges and four corners in the top layer rejiggered into their proper position. There are 21 PLL cases (22 if you count solved), each named after a letter that somewhat resembles the pattern of the edge and corner swaps. More detailed background and the PLL algorithms can be found on the speedsolving.com wiki.
A PLL time attack is the performance of each of the 21 PLLs in a stream without stopping. Completing it in a minute is respectable, but not amazing; forty-five seconds is really good; and better than that is truly impressive. When I first learned PLLs, I did a quasi time attack — quasi because it was truncated (I did the Gs separately), because I did each PLL individually (not in a single stream), and because I used my best of three attempts per PLL. The sum of the 21 parts was 66 seconds. Continue reading →
Yesterday, AL60Ri7HMi57 posted a quick video with new lighting and a clean 22-second solve. In the comments, she wrote, “Do the scramble and post your time in the comments below!” So, I figured I’d give it a shot. I had my camcorder charging on my desk, so I haphazardly aimed it, flipped it on, scrambled, and solved. 35.65 seconds. Not great. But, in all honesty, right about where I am. At least on video. I’m about 5-7 seconds faster off video, without the inexplicable nervousness of being on-cam.
The Video and Initial Observations
My first reaction was to ignore the solve and move on. Rarely one to miss an opportunity for self-examination, though, I decided to learn from it. So, for better or worse, here’s the video.
I remember learning the V Perm early in my PLL progress and being quite proud of myself. I even wrote this dedicated post containing a video showing my unimpressively slow execution. Unfortunately, the execution never got that much more fluid. Although it was an average speed perm for me when I did my (kinda sorta) PLL attack, it always was one of my clunkier ones.
z D’ R2′ D (R2 U R’) D’ (R U’) (R U R’) D (R U’) z’
No awkward re-grips and much smoother than the traditional approach. He bemoaned the z rotations; I was intimidated by all the D layer moves. Still, the fact that he could sub-1 the perm led me to believe that I could do it in 2. As this video shows, I’m close:
So, yeah, I’m still cross training, although I don’t expect to harvest any significant improvement for a while. In the meanwhile, I decided to learn a couple more OLLs that seem to crop up frequently — the single dot cases. The two cases (OLLs ##1 and 2 on the speedsolving.com wiki) are mathematically somewhat rare — with their probabilities at 1/108 and 1/54 respectively. That’s the equivalent of a 1-in-36 (or about a 2.8%) chance of having either of these cases arise during a solve. They seem to occur more frequently for me, probably just because not knowing how to one-look them made them more conspicuous.
I don’t usually post about each OLL or PLL I learn. But there seemed to be a dearth of tutorials on these two cases, with the few videos I found online lacking any technique focus or finger-trick explanations. Also, both of these OLLs seemed vulnerable to simplifying alternations to make them more accessible to beginners or novices. Here’s my short video tutorial:
It’s been 14 months since I wrote about more advanced cross techniques, which is only slightly less long than I’ve been at this cubing thing altogether. At the time, I was thrilled just to move on from the beginner’s method of forming a daisy on top. Memorizing the cube’s color scheme seemed like an accomplishment.
A year-plus later, I’m stuck at around 35 seconds. I’ve learned all 21 PLLs, have gotten pretty fast at two-look OLL (with a handful of OLLs one-lookable), and am competent with F2L (fast, but with lots of hunting still). But my crosses are still really clunky. Time to start focusing on the cross.
To set a benchmark, I did 5 Ao12s of just crosses. I spread out the sessions to make sure they were pretty accurate representations. Although there are a few stray bests, the 60-cross average was 6.75: Continue reading →
I discovered a couple months ago on the speedsolving.com forum that Dayan made a small run of Zhanchis in clear plastic around April 2011. They were prototypes, and only about 100 were made. A few vending sites, such as 51morefun.com and lightake.com list them, but as sold-out at this point. Given the rarity, they’ve been hawked on Ebay for over $800!?!
I’ve always dug clear products. Getting to see the inner-workings of intricate machines is fascinating. So, the chance of getting my favorite puzzle in a translucent model was intriguing. The rarity of it made it that much more so. But I wasn’t going to drop 8 Franklins for what is otherwise a $12 puzzle!?!
I eventually found someone on the speedsolving.com forum who was willing to part with a new DIY kit at a reasonable price. I received it a couple weeks ago and finally got a chance to assemble it. Here’s a video: