Date:     Thu, 5 Nov 87 10:26:29 EST
From:     Bernie Cosell <cosell@WILMA.BBN.COM>

I picked up a "deluxe Rubik's Magic" at Games People Play the other day.
It is a twelve-square magic. Has anyone solved this guy yet? My wife
has been hacking on it some and and has managed to run it from the
starting state (2x6) to the target state (as in the normal Magic, but
moreso), but not enough comprehension of it all yet to get all the
circle pieces in the right places, yet.

Well, /my/ wife solved it.

It seems to be more fun that the normal magic because if you ignore
the circles you can make a bunch of interesting shapes (the big-hollow-
square was neat to blunder into).

You bet! As a matter of fact the order-6 puzzle is so much more fun
than the order-4 that I am wondering whether higher orders might be
even more fun. In my opinion the order-4 cube was /less/ fun than the
order-3, and it's a pleasure to see a puzzle where bigger really is
better.

Jenny and I have a conjecture that if a given flat shape is possible,
a flat shape that is derived from the possible one by moving a single
square one step diagonally -- is impossible. There is probably a parity
argument lurking somewhere that can prove this.

Is a similar puzzle with triangular tiles possible?