eps

Purpose

Floating-point relative accuracy.

Synopsis

eps

Description

eps is a permanent variable whose value is initially the distance from 1.0 to the next largest floating-point number. It can be assigned any value, including 0. eps is a default tolerance for pinv and rank, as well as several other MATLAB functions, so changing the value of eps changes the way some functions operate. On machines with IEEE floating-point arithmetic, eps = 2^(-52), which is roughly 2.22e-16.

See Also

isieee, realmax, realmin

(c) Copyright 1994 by The MathWorks, Inc.