floating-point notation

System in which numbers are represented by means of a decimal fraction and an exponent. For example, in floating-point notation, 123,000,000,000 would be represented as 0.123 × 1012, where 0.123 is the fraction, or mantissa, and 12 the exponent. The exponent is the power of 10 by which the fraction must be multiplied in order to obtain the true value of the number.

In computing, floating-point notation enables programs to work with very large and very small numbers using only a few digits; however, it is slower than fixed-point notation and suffers from small rounding errors.