I'm working through An Introduction to ANSI C on UNIX which has suited me well so far.

I don't understand this problem they gave me, though.

"Write a quadruple-precision integer package with integer arrays and the arithmetic routines: qint_add, qint_sub, int_times, qint_quotient, and qint_display."

I know pretty well about precision, I think, but why not use floats for only four decimal places? The only meaning I can think of for this problem has already been built into C.