Question: Matrix evaluation differ on i386 and amd64 platforms?

THE FOLLOWING CODE 

restart;

A:=Matrix([[ a , b ], [ c , d ]]);

a:=1; b:=0; c:=0; d:=1;

A; 

produces differents results under MAPLE 16  linux i386 and under MAPLE 16  linux amd64

in the first case the last evalution has the following printed output:

Matrix([[a,b],[c,d]])

on the second machine the printed output is

Matrix([[1,0],[0,1]])

Does anybody has an explication; I thought that the "coorect behaviour was the first one since tables use last-name evalutation. But now I am puzzled.

 

Please Wait...