FEPS Single Precision Floating Point Relative Machine Precision Epsilon

Section: Base Constants

Usage

Returns feps, which quantifies the relative machine precision of floating point numbers (a machine specific quantity). The syntax for feps is:
   y = feps

which returns feps for single precision values. For most typical processors, this value is approximately 2^-24, or 5.9604e-8.

Example

The following example demonstrates the use of the feps function, and one of its numerical consequences.
--> feps
ans = 
  <float>  - size: [1 1]
 0.00000011920929  
--> 1.0f+eps
ans = 
  <double>  - size: [1 1]
 1.0000000000000002