Slashdot Mirror
Four-Dimensional Rubik's Cube Craziness
roice writes "Rubik's junkies and puzzlers will be interested in this software rendered four-dimensional
analog of Rubik's Cube. With over 1.75E120 possible combinations, it's
a mind bender. Free versions are available for both Windows and Linux, and
they even publish their source code for download. Solving it will get your
name listed in their Hall
Of Fame, and there is also a running competition for the most efficient
solution. To help get you started, you can check out a solution algorithm based
on techniques used to solve the popular three-dimensional version."
Comment removed based on user account deletion
Maybe somebody subjected you to one of my favorite old tricks. Take one corner off of a solved cube and rotate it so that the colors don't match the rest of the cube. Reassemble in this orientation. Presto: unsolveable cube.
Maybe it's because I read some quack's claim that the 4th dimension was time. In which case a 4D rubics cube would solve itself over time or be onsolvable because it rescrambled while you were trying to solve.
The 4th dimension is whatever you define it as. There's no "true" 4th dimension. Einstein had the idea that time could be treated in a fashion similar to that of the spatial dimensions, and so, in his work, he defined the 4th dimension as the temporal dimension.
However, for this program's purposes, it's assumed that we're talking about the 4th *spatial* dimension. You can (kinda) visualize it if you think of the progression of first 3 dimensions: a line (1st d) can be rotated 90 degrees to itself to form a square (2nd d) which can be rotated 90 degrees to itself to form a cube (3rd d). The 4th dimension is thus hypothesized as the space defined by a cube rotated 90 degrees to itself.
(Granted, I'm no mathematician, so if someone has a better understanding, please correct me.)
This puzzle uses the first model mentioned above, except that you can only see seven cubes at once (the outer cube is hidden so that it won't block the view of the others). If you rotate the model (with Shift-left or Shift-right click), the outer cube comes into view.
Disclaimer: The opinions expressed are not necessarily my own, as I've not yet had my medication today.
The best way to solve a given cube is called "God's algorithm" to us Rubik's geeks (I can average about 50 s. solving a cube, which is ok, but not phenomenal). It's 18 moves. The method behind it is far too complicated for a person to do without a computer to assist, but using a rather simple method, a person can very easily solve a cube in around 65 moves, but slowly. The record-breaking solution times are closer to a hundred moves, but rather than remembering a move-efficient but thought-intensive way to solve it, one remembers many more algos that whose situation can be recognized much more quickly.
And for the previous posts asking how long it takes a computer to do it... it's very, very low. Under a second. Many people can do it, manually (a computer just has to give the moves, it can ignore the time required to actually turn the cube) in under 20 seconds (For the people out there in disbelief, Dan Knights has a video of him actually doing it in 17, it's for real. I won't post the link, because I'm not going to be responsible for slashdotting his site).
There's a huge difference between a computer solving it "brute force", and a person or computer solving it through established algorithms. By brute force, just twirling the cube until the solution popped up, it would take on average however long it takes your computer to process half of the possible combinations. That's quite a long time. However, a computer solving a cube how we would, focusing on time rather than least amount of moves, could easily solve more than one a second.
Puzzled by the cube? Try renting two (relatively low-budget, unknown) sci-fi flicks. - Cube: Buncha people, trapped in a buncha cubes, with a buncha deadly traps. - Cube 2: Hypercube: Buncha people, trapped in a hypercube, with less deadly traps but more confusion as to wtf is going on. Both movies are fairly puzzling in their own right, with that sort of "unknown" sci-fi ending that is commonly found in lower-budgeted movies (e.g. Pi).
http://www.chmodoplusr.com/
Douglas Hofstadter wrote a couple of excellent columns on Rubik's cube and variations on the theme for his Metamagical Themas column in Scientific American back in the eighties (you can buy his collected columns in this book). In particular, he talks about the various ways you can modify the basic 3x3x3 cube concept - for example, 4x4x4 cubes, 3x3x3 tetrahedra, alternate colour schemes, and so on (along the way, investigating the spark of inspiration that encourages people to try out different variations on a theme - something he refers to elsewhere in his books as 'conceptual slippage' - this hypercube would be a 'slip' along a different axis to those hofstadter explores - I'm sure he'd appreciate it :) ). He goes into plenty of detail about the mathematical approaches you can use to solving the cube, and some intriguing analogues to subatomic physics that crop up in the maths of rubik... anybody wanting an introduction to the kinds of topics the people behind this hypercube are exploring could do worse than to read those articles.
There's also some excellent stuff in that book on Lisp, quantum mechanics, chaos theory, Alan Turing, and nuclear war... great selection of articles by an extremely interesting mind.