Rubik's Cube

So I bought a cube and I solved it.  At times I was tempted to read the little booklet or look online for clues but, ultimately, I wanted to prove that I could do it unassisted.  6 years ago I had a cube and I learned the supplied solution. Fortunately, since that time I had completely forgotten how to solve it.  All I could remember was that the solution suggested solving in layers and that the last layer was a real bitch.

Things I did right:

  1. I devised a simple system of notation.  This was the first thing I did and the thing that paid off the most.  I drew a diagram of a cube and labeled each possible rotational layer: ABC across the top left, DEF across the top right and GHI down the side.  Basically, I started playing with and recording simple patterns that returned most of the blocks back to their original positions.  An example of such a pattern is: 2G, 3A, 3G, 1A.  The number indicates how many 90 degree twists to give that layer – 4 twists would turn it all the way around and do effectively nothing, 3 twists is the same as twisting it backward -1.  I soon discovered that the only important layers to twist were A, F and G.  I experimented a little with H, but was unable to find anything promising.
  2. I developed a visual key.  I honed this over the course of my 3-4 weeks or so of solving.  Initially, I would draw a tiny cube and place a tiny circle at each location where a block changed, and draw arcing arrows between them to show the direction of movement.  I soon began coloring in black those blocks that rotated their facing as well.  Later, I was adding a little number to indicate if the rotation was 1 spot clockwise or 2 (this was only for corner blocks which have 3 faces).  I also began to draw the results of combo patterns that diagramed combinations of, or repetitions of, the simpler patterns.  These combinations ultimately proved to be essential to finding a solution.

Things I did wrong:

  1. I Wasted time trying to get results with limited ‘tools’.  Because I did not know the scope of the project, I would all too often stop my research and attempt to complete the cube with the limited number of patterns that I had documented.  In some ways this was useful because it led me to understand very clearly the problem of position vs. facing – blocks may be in the right place but be facing in the wrong direction.  Eventually I got to the point where I could theoretically solve the puzzle, but the few useful patterns I had were much too long and I invariably made some wrong twist along the way and had to start over.  I wasted a lot of time before I gave up and returned to researching more patterns.
  2. I wasted time redrawing my charts.  I started work in a notebook, but the many tiny diagrams became a jumble so I painstakingly organized and drew them out on a large sheet of paper.  I later reorganized them and inserted new patterns and I had to redraw the whole sheet.  Eventually I decided it would have been best to put each pattern on a ‘flash card’ of sorts so that I could experimentally rearrange them and view any two, immediately, against each other.
  3. I was unable to find a simple, corner-spinning pattern (when 3F3G1F2G3A3G1A x15 is your best option, you’re in trouble).  I had a big problem with determining the order of the following 4 necessary actions to completing the third and final layer: place sides, place corners, spin corners, spin sides (spinning refers to re-facing the blocks).  I finally decided on the previous order when I discovered a moderately simple pattern that allowed me to spin the sides without disturbing any other blocks.  I gave myself no real direction in solving this order and only in retrospect (and a second attempt at repeating the solution of the cube) have I realized that my biggest problem may have been solved by committing further research to finding simpler patterns for spinning the corners.

In the end, I learned a lot about the cube. 

  • I don’t think there is a person alive who could pick up a cube and figure out how to solve it without writing anything down – the memory required would be excruciating. 
  • In order to solve a Rubik’s cube one must be able to draw lots of very clean little cube diagrams.
  • People who speed-solve the cube are split into 3 categories:

§          Idiots who think they are clever because they memorized the patterns in the booklet and realize that most people will assume that they are brilliant

§         Nutcases who solve the cube on their own, then optimize the solution and then, rather than cast the cube away, absolutely sick of it, they memorize the patterns because they realize this is the only way they will get any credit for all of their hard work

§         Brilliant savants who memorize such a plethora of alternate strategies and are able to implement them in such immediate circumstances that they must have a brilliant understanding of the cube.  These are the people that can solve a cube from any orientation in under 60 seconds.  Wow.

  • Erno Rubik himself took nearly an entire month of concentrated effort to solve the cube for the first time.  I kicked his ass.

 

 

 


about | home