By all accounts, the Rubiks Cube is cool again. I found this out accidentally from my students. Having a Cube on my desk (an accessory to my "mad scientist" persona) is an actual magnet for them. And although I'm not technically supposed to, I will let the occasional attention-deficient type of kid Cube rather than tear around the room and start a riot, or play with annoying and prohibited personal electronics. I'll even let a Cube walk out the door now and then, it doesn't matter, they generally come back eventually.
And the kids seem unaware that the Cube was a 1980s thing. They think it's new and cool. A few have scoffed at my contention that I learned it when I was younger than they are, how could I have learned it when it wasn't invented yet? Poor kids of today, surfing endlessly recycled waves of culture. It isn't just here - I hear from a reliable source that Rubik's Cube is huge in India right now, of all places - Indian yuppies sending their kids to Rubik's classes, the mind simply reels.
Given all this unexpected, intense interest in Cubes and having Winter Break ahead of me, I felt ashamed to be Cubing at a 3 or 4 minute timescale, so I ventured in search of newer and better solution methods. The days of Nourse's "The Simple Solution to Rubik's Cube" being long gone, I started with the instructions in the box. Remember when the Cube did NOT come with instructions? Life was cheap back then, wasn't it? So your Dad played with it for a day or two, got really mad, drove back to the store, and bought a copy of Nourse.
Good old Nourse. Do the top edges ("the cross" in modern lingo), do the top corners, do the middle edges, get the bottom corners in the correct positions, twist the bottom corners into correct orientation, get the bottom edges in correct position, orient the bottom edges. A solution which even to a casual eye starts off okay and just gets slower and slower as you go. Orienting the bottom edges could easily take as long as doing the whole rest of the puzzle combined, using Nourse's method. And woe is me, I have committed his ungainly sequences to indelible muscle memory and they sing to me in a voiceless voice "Elminus arplus efplus elplus arminus beeminus. Elminus arplus efminus..." Yet I will never do a 60 second Cube his way and neither will you.
There was another well-circulated guide back in the day, now out of print - Minh Thai's corner-first solution . It was hyped as much faster, and probably was, provided that you memorized about 50 or 100 or god knows how many sequences. I still remember a few of his - they help a little with the dog-slow Nourse method, but mostly were handy for making patterns and such. But was Thai the guide to a 60 second Cube? Not for me, I think. If it was meant to be, I'd have gotten it 25 years ago.
Now I'm a middle-aged married man in my mid-30s and like generations before me I groan that the kids these days sure have it soft. We have become a mature, risk-averse culture and one of the consequences is that people can't handle failure. So now, the Cube comes with a nice easy solution guide right in the package. How good is this solution? Instead of just tossing the pamphlet at my students, I actually read it. Then I checked the Rubik's website and found a less-simplified, far more efficient version of this Cube solution method.
It starts exactly like Nourse. Do the "cross". Do the top corners. Do the mid-edges. This really isn't so bad, though. Nourse's solution started off quite well, it was the ending that sucked you into endless quicksands of time. The ending of the solution, though, is a revelation. Back in the day it was a cliche - novice Cubists were mocked for putting random cubelets with one common color on a common face together and thinking they'd made real progress - even Homer Simpson once boasted, "I got a SIDE". It's not a joke anymore. That's actually how you do the last face - you flip all the last-layer corner and edge cubelets so that the correct color is up, even though they're out of position. Then you permute first the corners and then the edges, with some very clever sequences, and before you know it you've solved the Cube. The new sequences are much shorter than Nourse's. And yet, they do not reorient the last-layer cubelets at all. Start with orange up, they finish with orange up, every time. This is amazingly fast and elegant
compared to Nourse's long, long stretches of "elminus arplus".
The Internet tells me that the new technique is called "OLL/PLL" as opposed to the old method of "PLL/OLL". Orient last layer, permute last layer. I love it.
There has also been a change of point-of-view. While I will probably always think of the first of the 3 layers to solve as the 'top', it's no longer considered the top. Pretty much everyone now calls it the 'bottom'. No longer are we to start solving the Cube from the top and proceed to work 'down', because there's no point in looking at the solved portion of the Cube! Your eyes should be on the scrambled, unsolved portion. So you don't push cubelets up, you shove them down. As once written in another 1980s comeback, due back again at a theater near you this summer, "the enemy's gate is DOWN". Not only is this bottom-up approach better for inspecting the Cube, but it leaves the important faces free for fast manipulations known as finger tricks. Someday, perhaps, I'll learn how to do those. A good solution lets you solve the Cube in less moves - but making the moves faster also is important to a competitive time.
Pretty soon I was chugging through the cube in 2 minutes or less. A little of the wife's sewing machine oil had my Cube spinning like a racing bike. But I wasn't there yet. A bit of Internet research revealed that the all-time popular algorithm among speedCubers was called "CFOP", or "Fridrich's Method". The only real difference between CFOP and the one on Rubik's site is that the top corners and mid-edges are combined into a single stage termed "F2L", for "First Two Layers". This brought back memories of Nourse, who hinted that combining these stages was key to competitive solving. Nourse knew! Fridrich's method dates back to the 80s itself. But Fridrich used OLL/PLL to finish the Cube. When using the Nourse method, there was no incentive to optimize F2L because PLL/OLL was so slow at the end. Almost as if Nourse deliberately refused to publish the high-end methods. I suppose there used to be an ethic about magicians not revealing tricks, perhaps the same ideology was active during the early days of the Cube. Anyway, there are about 75 Fridrich cases where one corner cube and one edge cube, in a particular position with a particular orientation, can be simultaneously placed. I'm not about to memorize 75 sequences.
But just knowing that it's possible is enough to help me develop an intuition about how to assemble and place top corner-midedge pairs, transcending memorized algorithms and just putting the cubelets efficiently where they belong. And doing this, I've gotten myself into the 75 second average range. Still not regularly below 60, but I feel like I've accomplished something these past couple of days. A good F2L now makes me feel like a Zen master. To take what the Cube gives me and put it together on the fly, it's a great feeling. Run through the OLL/PLL on autopilot, it's easy, those rare sub-minute solves let me feel that even if I never got to be a famous scientist or a rock star or something, I could still show young Me a trick or two.
There are other methods out there - perhaps I'll learn Petrus or even Heise - but this is good for now.
I believe my old 4x4 "Rubiks Revenge" is packed in a box somewhere a thousand miles away from here, so I might as well just order another off of Amazon. And my 5x5 Professor's Cube, peeling, long-ignored - its stickers are generally known to be defective as manufactured. Laughably, they consist of plain white stickers with squares of colored vinyl clinging to them by static electricity alone. Of course they didn't last. New ones are on order from Cubesmith, and I'm looking forward to refurbishing it and Cubing large once again.
As for my students - a couple of them can pretty reliably do and undo Checkerboard and Six Boxes, and that's pretty much all they can do - solve it? No way! I'd teach them more, but they really need to learn actual science in my room. That's why we're actually there, after all. If they can't figure out how to Cube on their own time, no sympathy. It's easier than ever to learn. Kids these days.
![]](images/rt_edge3.gif){kind=small}
![[XML]](images/rssblue.gif){kind=small}
![[P]](images/print3.gif){kind=small}
