Site hosted by Angelfire.com: Build your free website today!
Table of Contents Previous Next

1.4 <float.h>

The names in the table below, a subset of <float.h>, are constants related to floating-point arithmetic.  When a value is given, it represents the minimum value for the corresponding quantity.  Each implementation defines appropriate values.

FLOAT_RADIX 2 radix of exponent representation, e.g., 2, 16
FLT_ROUNDS floating-point rounding mode for addition.  The values can be:
   -1  indeterminable
    0  toward zero
    1  to nearest
    2  toward positive infinity
    3  toward negative infinity
All other values characterise implementation-defined rounding behaviour
FLT_DIG 6 decimal digits of precision
FLT_EPSILON 1E-5 smallest float number x such that 1.0 + x not equal 1.0
FLT_MANT_DIG number of base FLT_RADIX digits in mantissa
FLT_MAX 1E+37 maximum floating-point number
FLT_MAX_EXP maximum n such that FLT_RADIXn-1 is representable
FLT_MIN 1E-37 minimum normalised floating-point number
FLT_MIN_EXP minimum n such that FLT_RADIXn-1 is a normalised number
DBL_DIG 10 decimal digits of precision
DBL_EPSILON 1E-9 smallest double number x such that 1.0 + x not equal 1.0
DBL_MANT_DIGIT number of base FLT_RADIX digits in mantissa
DBL_MAX 1E+37 maximum double floating-point number
DBL_MAX_EXP maximum n such that FLT_RADIXn-1 is representable
DBL_MIN 1E-37 minimum normalised double floating-point number
DBL_MIN_EXP minimum n such that FLT_RADIXn-1 is a normalised number
LDBL_DIG 10 number of decimal digits, q, such that any floating-point number with q decimal digits can be rounded into a floating point number with p radix b digits and back again without change to the q decimal digits.
LDBL_EPSILON 1E-9 smallest long double number x such that 1.0 + x not equal 1.0
LDBL_MANT_DIGIT number of base FLT_RADIX digits in mantissa
LDBL_MAX 1E+37 maximum representable finite floating-point number
LDBL_MAX_10_EXP +37 maximum integer such that 10 raised to that power is in the range of representable finite floating-point numbers
LDBL_MAX_EXP maximum integer such that FLOAT_RADIX raised to that power minus 1 is a representable finite floating-point number
LDBL_MIN 1E-37 minimum normalised positive floating-point number
LDBL_MIN_10_EXP -37 minimum integer such that 10 raised to that power is in the range of representable finite floating-point numbers
LDBL_MIN_EXP minimum integer such that FLOAT_RADIX raised to that power minus 1 is a normalised floating-point number
Table of Contents Previous Next

Last modified: Mon 14 Dec 2000