[gecode-users] magic squares

[Marco Correia]

hi,

I was playing with the magic-squares example, and found that the first 
solution to the problem when using INT_VAR_NONE and INT_VAL_MIN as heuristics 
(and using no symmetries) is

8     9    6    11
12  14   3    5
1     7   10   16
13   4   15   2

instead of 

1   2    15 16 
12 14  3   5 
13  7   10 4 
8   11  6   9

why?

ps: I also changed the line 116 of magic-square.cc to m(j,i) to make it print 
lines as lines and cols as cols.

thanks
Marco


-- 
Marco Correia <mvc at di.fct.unl.pt>