The term arithmetic underflow or simply underflow has the following meanings.

1. In a digital computer, the condition that occurs when a calculation produces a result that is less than a given register or storage location can store or represent.

Underflow is analogous to overflow except that the magnitude is too small rather than too large.

For floating-point arithmetic, an underflow occurs when the magnitude of an intermediate result (i.e., before rounding) is strictly between the smallest normal floating-point number in the format and zero. As specified in IEEE 754 the underflow condition is only signalled if there is also a loss of accuracy. Typically this is determined as the final result being inexact. However if the user is trapping on underflow, this happens regardless of consideration for loss of precision.