MATLAB Function Reference

realmax

Largest positive floating-point number

Syntax

• n = realmax
  

Description

n = realmax returns the largest floating-point number representable on your computer. Anything larger overflows.

realmax('double') is the same as realmax with no arguments.

realmax('single') is the largest single precision floating point number representable on your computer. Anything larger overflows to single(Inf).

Examples

realmax is one bit less than 2¹⁰²⁴ or about 1.7977e+308.

Algorithm

The realmax function is equivalent to pow2(2-eps,maxexp), where maxexp is the largest possible floating-point exponent.

Execute type realmax to see maxexp for various computers.

See Also

eps, realmin, intmax

reallog

realmin

© 1994-2005 The MathWorks, Inc.