A cube…

Rubik's Cube: the first generation

Some memories going back over 35 years

Rubik's cube sequencesI was given my first cube as therapy at the age of 14, as I had broken my wrist in Judo and people feared for my violin playing. It appears that I did not take to it at first (probably because I saw no hope of solving it) but my mother was quite keen and eventually we discovered that it was solvable: indeed, that somebody at school could do it in less than ten minutes! We beseeched him for help and he restored it to Start (the pristine condition). I can still remember the sense of total wonder as we saw it return to the perfect form; perhaps it was the first time in my life I had experienced true Magic, Beauty, and Grace. We had managed to corner the "solver" on his last day at the school, and resolved to exploit this opportunity to the ultimate — we would methodically explore one sequence of moves after another until we had a comprehensive understanding of the whole device. Little did we know.

I also remember quite clearly how we sat in the kitchen that afternoon. The summer holidays had begun, so I had all the time in the world and a complete Cube to boot. We resolved to try working back to Start from easy positions, just a few turns away. My mother did 3 turns; I was quite sure I could tell what she had done and tried to reverse the 3 turns; I failed, so undid my 3 turns; she then undid her original 3 turns. The Cube should have returned to how it was. In theory. It didn't. We sat there for hours.

That was when the madness started. So near but yet so far … As the Barron Knights said in their song, "I just twist your cube all day — this ain't my idea of child's play". I filled notebook after notebook (see example to the right — all sequences had names), had sheets specially printed with expanded diagrams of parts of the cube to save time when noting down new operators, and abandoned almost all other interests to the exclusion of the Cube. On the bus to school, on the bus from school, in lunch breaks, and even under the desk in lessons, I was twiddling. In one sense it bore results, as I developed a greater repertoire of moves than anyone else I encountered — several hundred (compared with around 8 for most people). Incidentally, my method is none of those listed at the speedcubing wiki — but it is quite likely that many of my algorithms have since been (re)discovered and added to the algorithm database.

Fame & Fortune

It had to come: the National Championships, in December 1981. I entered the London heat, held in the Penta Hotel tower block, but did not think I had much chance of winning. Although my best time at that stage was 35 seconds (I did achieve 30 seconds when warming up on the day), a friend called Richard who was also taking part could do it in under 30, and even Nick Hammond (one of the first cubemeisters to publish a book in the UK) was due to be there. When I arrived I found that I was only number 11 on the list … with number one having registered a best time of 5 seconds! I still turned in a reasonable time of around 46.12 seconds and waited to see it fall. But Richard seemed to panic due to the media attention, and returned his "finished" cube one move too early; Nick Hammond too was a couple of seconds outside (although he had already qualified in a previous heat with a time of 35.38 seconds); and the Number One contender had apparently only achieved the incredible 5 seconds once by fluke, with his normal time being over 2 minutes. So I ended up winning the heat and going on to the final (see press cutting below).

Press cuttingThe final was even grander — it was held in the River Room at the Savoy in the presence of Ernő Rubik himself, as well as other cubology celebrities such as David Singmaster (a mathematics professor at the University of the South Bank who has written prolifically about the cube and group theory). The evening before the competition we were given a night out at The Talk of the Town (where our glasses were never allowed to be empty) and were even taken backstage later to meet the star of the show, Anita Harris.

The next day the ten finalists assembled. Each of us had brought our favourite "racing cube", some lubricated with graphite, others with WD40, vaseline or car grease. I remember one contender from Ireland — who could "undo" any 10 turns on a pristine cube — adamantly refusing to remove his gloves in case his hands got too cold and became stiff. Another one wore sweatbands, although I suspect they were no more effective than go-faster stripes. All the cubes were handed in and mixed randomly, and stored in boxes marked with our names. They were then placed on an automatic timer, and one after another (in order of qualifying times) we solved them and slammed them down. My final time was 33.60 seconds, which gave me fifth place (first place was taken by Julian Chilvers with 25.79 seconds; he went on to the world final in Budapest, which was won by the Vietnamese American Minh Thai in 22.95 seconds).

After the contest we finally were able to relax and exchange strategies and sequences. It turned out that most people had been largely relying on quick hand movements and had assumed, when they saw how slowly I was moving, that I had a bad hangover from the previous night! Some of them were actually turning the cube over four times faster than I was … I was able to "catch up" purely on the basis that my method was much more efficient in terms of numbers of turns. Of course, Julian won because he had a good method and fast hands.

Epilogue — the Revenge

A couple of years later we were all contacted again regarding a national competition for the Rubik's Revenge, i.e. the 4x4x4 cube. I immediately said I would take part, but in the event only a few of the other finalists did. The cube craze had largely faded away — and nothing could have shown it better than the switch from the Savoy River Room (top prize: a trip to the world championships) to the third floor of a department store in Romford (top prize: a Ford Fiesta)! There were no heats, and the final took place on June 25, 1983. The organizers also appeared to know less about our art, as not only did we have to use brand new cubes (which were stiff, and above all slippery), each of the cubes was also arranged in exactly the same pattern (which gave the last person in each round an advantage — some might think this would make it fairer but actually such considerations are irrelevant due to the variety of methods and each person's preferred colour scheme). The only other person I knew there was Julian Chilvers. Not surprisingly, his best time was already much faster than mine and he won in under two minutes, while I came second.

The years since

So was this the end of my cubing? Not really. I have not bought every "Rubik" invention since, but I have bought a few and each of them can keep me occupied for hours on end. I don't think I could ever be bored with a cube unless I discovered "God's Algorithm", which is hardly just around the corner (or it wasn't when I first wrote this… since then, UCLA Professor Richard Korf and then Kent State University Professor Morley Davidson have respectively managed to produce a method for deducing algorithms for solving any cube in an optimal way, and also demonstrated that the most complex positions require 20 moves to solve, although the mechanical nature of the computations mean it doesn't feel how God would solve it: they can't explain why each turn is done, they just know the result, rather like many cubists who simply memorize sequences). For years I hoped some day to enter all of my operators into a computer and work on better solutions, but by the time suitable programmes had been written, I was too busy earning a living — and others had taken speed cubing to new heights, so my work would have been rather superfluous, no more than a historical footnote (update: I have now started doing this… yes, probably superfluous, but still satisfying; it is still surprising that nobody has published routines to find all the variants of a sequence, although it should be quite trivial to code; Kociemba's Cube Explorer is handy, in that it detects when two sequences produce isomorphic results, but it doesn't allow fuzzy facelets, for the times when you don't care what happens to particular cubies). I still find the combination of a two-dimensional computer screen with input methods such as a keyboard or mouse, or even voice recognition, quite primitive when compared with what the human eye, brain and hand can achieve with a physical cube. If they invent a robot that can complete a normal cube in anything near the time a human can, in my lifetime, I will be amazed (OK, they've now done it, sort of… but it doesn't say if the time includes everything, e.g. image acquisition and computation — and indeed, the setup time is likely to be 100x longer than that for a human, who could simply catch a cube thrown to them and solve it within a few seconds).

As Douglas R. Hofstadter wrote in his book "Metamagical Themas", the cube "is more than just a puzzle. It is an ingenious mechanical invention, a pastime, a learning tool, a source of metaphors, an inspiration … It furnishes analogies to particle physics, biology, problem-solving in everyday life, entropy, path-finding and even theology". I find the cube humbling because you are always aware of how imperfect your methods are, but at the same time encouraging as any progress you make is quite obvious. It is certainly beautiful — a wide variety of patterns is possible even on a basic 3x3x3 cube, and far more can be achieved on the larger models. I would love to create a mosaic picture with a 20x20x20 cube, but not only has nobody made one this large yet (oh yes they have!), the time taken to achieve anything on the larger sizes of cube appears to increase exponentially so I would probably die in the attempt (nope, even a 17x17x17 only takes 7.5 hours); still … I also enjoy the cube because of the balance it requires between physical and mental dexterity, and on a deeper level the balance it can require between the realms of logic and reality — for example, no human can explain what every single twist they make achieves, but if they did not have some idea of what each sequence achieved they would not be able to solve it. Sometimes I find myself distracted halfway through a move by cubies fortuitously falling into patterns that "feel" right — another aspect that would be hard to code for a computer. I'm sure it would be possible to derive some theories about artificial intelligence from this, and maybe it has already been done (some say cube-solving programmes demonstrate AI, but that's not the kind of responsive/adaptive 'thinking' system I mean).

I actually see many similarities with the martial arts. It is not simply the need for problem-solving, which is one side of the martial arts (and of my work as a martial arts translator) that I really enjoy, but also the recognition that logic on its own is not enough to give it life and turn it into something of beauty, an art; the striving for perfection despite knowing that perfection could only come through divine inspiration (and that conscious striving actually blocks out such inspiration); the ability to keep playing without any need to reach some specific objective; the need to internalize moves to be able to use them, and the need to bypass conscious thought to be able to do so quickly; the contradictions and confusion and connections and conundrums and controlled chaos.

Finally, I enjoy the humour that one can generate with a cube — it is not merely young children who become spellbound when faced with a cubemeister's magic! No matter what the Barron Knights might say, I still think it's a lot of fun. (And now, 35+ years since I got my first cube, I am still discovering new things and sharing them with others. I doubt I'll ever get sub-10 — with aging fingers & eyes, even sub-20 would be a challenge — but I recently bought a timer so we'll see… -> and a few years further on, I was given an Internet-connected GoCube — on my first attempt I was delighted to get 35.286s, just 1.6 seconds slower than my 1981 national final time; but now it only makes me 5242nd worldwide, so a lot of work would be needed to get within the top 1000: even Minh Thai's first title would be far outside.)


When I first created this page (circa 1990? the bits in italics etc. are post-millennial edits) there were very few cubic links on the Web (fewer than 3000 websites in total in 1994 — and the first proper search engines were yet to appear), but now Google shows over 8 million. Most are rather boring, though — shops, lists and videos of records, 32,000+ descriptions of the Fridrich method… Although it's old, the List of Rubik's Cube Resources might have a few of interest, and RedKB's "unboxing" videos can be entertaining. Cyotheking also has some more up-to-date links, and sets of algorithms. From 2016 onwards I've discovered a lots of information on 4-dimensional cubes, 5-dimensional cubes and the like, using software like Magic Tile — interesting and challenging, although rather skewed towards mathematicians.


This page produced by Ben Jones. All rights reserved.