As a test of my programming ability, I tried to write a
little program that inputs the diagonal of a monitor and
calculates the viewable area. It then optionally inputs
the diagonal of another monitor, calculates the viewable
area, and compares the two areas. (I got this idea when I
upgraded from a 15-inch to a 17-inch).

My algorithm is based on the Pythagorean Theorem and the
fact that monitors seem to have a 3:4 height-to-width
ratio. After a cin statement to get the float variable
inputDiag, which is the monitor's diagonal, it goes like

diagSquared = inputDiag * inputDiag;
heightSquared = (diagSquared / 25) * 9;
widthSquared = diagSquared - heightSquared;
height = sqrt (heightSquared);
width = sqrt (widthSquared);
area = height * width;

return area;

My questions are as follows:

1. Is there a way to improve this algorithm?

2. If the values 25 and 9 above were changed to variables,
would the algorithm, as a function, be a candidate for
coversion to a header file I could later #include in any
programs that might need it, just to save me having to
type it in all over again?

Thanks for all replies.