I've looked everywhere for any programs that handle Rubik's Magic... (not to be confused with Rubik's Magic Cube as the cube is sometimes called).
Rubik's Magic is the 8-tile (or sometimes more) flat puzzle that's connected by nylon wires... It allows folding on various axes. See
Of course representing it in a program is the first challenge. Which transformations are logically possible, or can be physically achieved without stretching the wires too much (perhaps a parameter of how much percentage of stretch is allowed?) etc. etc.
Then showing the puzzle on the screen would be another fun challenge.
Anybody have any thoughts?
That would be a tough one - much more complex than the cube because with the cube all of the moves can be done at anytime - with the Magic there would be far more constraint. If I had one in front of me I'd just play with it until some obvious patterns as to what you can/can't do and when emerged - then formalise them. But I can't remember the puzzle that well (it's been many years since I played it).
Maybe you should try Rubik's Clock next - it's a smaller step up from the cube.
Yeah. I thought of just exploring all possible paths and writing them down... encoding them eventually. Hopefully some pattern would emerge that would allow massive logical shortcuts rather than ending up with a huge lookup-table.
But would I have ever stumbled upon that weird diagonal transform the speed-solvers use as seen on YouTube.
The other way, and I'm leaning towards this one, is to actually represent the tile movement legalities by modeling them in some 3D internal representation. Not only the way the wires restrict movement, but also in some cases physical interference would prevent some moves.
It looks to be quite a brain teaser on napkins during coffee breaks at work.