Fog Creek Software
Discussion Board




Twisted Wires

Here's a nifty puzzle from Clifford Pickover:

http://sprott.physics.wisc.edu/pickover/puzzle.html

There are several puzzles there, but for now I'm referring to the Twisted Wires.

Can you do it?  (He's right; it is solvable.)

Paul Brinkley
Wednesday, February 27, 2002

Yes.  Posted a picture of it solved here:
http://www27.brinkster.com/theschnitz/connect.gif
If you're not getting it, keep trying.

Dave Schnizlein
Wednesday, February 27, 2002

my boss went away early today. that prompted me to pick this one up and solve it (in 20 minutes).

shailesh kumar
Thursday, February 28, 2002

Wow, I don't claim to be a genius or anything, but I solved this one in about 5 seconds.  Am I missing something...


SPOILER ****

Draw a straight line from C to C.  The lines from A to A and B to B just go around the C box at the top.

Chris Rickwood
Thursday, February 28, 2002

Way to go, Chris.  It took me a minute or two.  Part of me was conditioned to believe it was unsolvable, because I had seen a similar problem many years ago that -was- unsolvable.  Finally I got it by proving that two particular configurations of the B-B wire were impossible.

Sounds like you just happened to first try the thing that would make the solution obvious.  :-)

Paul Brinkley
Thursday, February 28, 2002

An interesting puzzle.  For the first ten seconds that I looked at it, I was stymied by my brain refusing to believe that there could possibly be a solution to this problem.  I couldn't get past the thought of "it can't possibly be solvable, so why even try?".  I was very close to giving up on it before I even started thinking about it properly, which somewhat concerns me.

Anyway, as luck(?) would have it, I ended up going "since I'm here, lets just look what happens if I connect this box to that one round the back", and then the solution sprang to mind within a whole two seconds of thought.  I feel I should learn something from this experience, although I'm not quite sure what... :-)

As well as Chris's solution previously mentioned, you can create a "mirror image" solution around connecting A to A directly rather than C to C.  But curiously, my solution was also based around the direct C to C connection.  Is that usual?  Is it anything to do with being right handed?

Carl Reynolds
Friday, March 01, 2002

i think perhaps the larger amount of space around the upper-left C block is why we (unconsciously) round the two wires around it, instead of around A (with limited space)

Dave Schnizlein
Saturday, March 02, 2002

This was a fun puzzle, which took me about 10 minutes.  My question is, why was it so hard for those scientists? 

(according to the article)  "About twenty percent of the scientists surveyed said this problem was impossible to solve. "

James Carter
Thursday, March 07, 2002

This took me a few minutes.  When I figured out the solution then it looked obvious.

Most of my time was spent just muddling around trying to figure a way to approach the problem.  When I realized the it had left-right symmetry I started by connecting B-B directly.

The direct B-B connection made it impossible to connect either A-A or C-C.  A little modification to the path made it possible to connect A-A.  It took a few seconds to realize that, of course,  I could extend the concept, move the B-B path around the upper C box and leave the plane divided into two areas with both A boxes in one part and both C boxes in the other.

mackinac
Thursday, March 07, 2002

Ya it didn't take long to figure it out, but I think it's because we were told that it was solvable. If not I probably would've said it was insoluble topologically. BTW, I did the straight C-C too: right handed.

Norman Spears
Tuesday, March 26, 2002

*  Recent Topics

*  Fog Creek Home