The reason for the calculator's behavior is the fact that you are trying to do decimal math on a binary processor. Not all decimal fractions can be represented exactly in binary form. This leads to very small errors in calculations and is demonstrable in most computers.
The way around this is to keep the error to an insignificant amount and round the result to desired number of decimal places. For example, if calculations are being done in dollars and cents, only two decimal places are needed. So when the calculator reports that 5.00 - 4.90 - 0.10 = 8.80914E-20
(0.0000000000000000000880914), the result is rounded to 0.
There is an article on page 42 in the February 20, 1995 issue of MacWeek that
also explains this behavior.
Support Information Services