Well, if the values are of a floating point type, then yeah, there would be a problem with those expressions since they won't be using integer division. But if they are of an integer type, then there is no problem since this is just a matter of mapping one integer to another based on the mathematical pattern, i.e., the inner indices vary one by one, then the middle ones vary three by three, then the outer ones vary nine by nine.Originally Posted by SirPrattlepod