The float.h header file consists of macros that specify the characteristics of floating point number representation for float, double and long double types.

• These macros are listed in the listing "Floating point characteristics"
Compatibility:

ANSI

BeOS

EMB/RTOS

Mac OS

Palm OS

Win32

"Floating point characteristics" lists the macros defined in float.h. Macros beginning with FLT apply to the float type; DBL, the double type; and LDBL, the long double type.

The FLT_RADIX macro specifies the radix of exponent representation.

The FLT_ROUNDS specifies the rounding mode. Metrowerks C rounds to nearest.

Macro	Description
FLT_MANT_DIG, DBL_MANT_DIG, LDBL_MANT_DIG	The number of base FLT_RADIX digits in the significand.
FLT_DIG,   DBL_DIG,   LDBL_DIG	The decimal digit precision.
FLT_MIN_EXP,   DBL_MIN_EXP,   LDBL_MIN_EXP	The smallest negative integer exponent that FLT_RADIX can be raised to and still be expressible.
FLT_MIN_10_EXP,   DBL_MIN_10_EXP,   LDBL_MIN_10_EXP	The smallest negative integer exponent that 10 can be raised to and still be expressible.
FLT_MAX_EXP,   DBL_MAX_EXP,   LDBL_MAX_EXP	The largest positive integer exponent that FLT_RADIX can be raised to and still be expressible.
FLT_MAX_10_EXP,   DBL_MAX_10_EXP,   LDBL_MAX_10_EXP	The largest positive integer exponent that 10 can be raised to and still be expressible.
FLT_MIN,   DBL_MIN,   LDBL_MIN	The smallest positive floating point value.
FLT_MAX,   DBL_MAX,   LDBL_MAX	The largest floating point value.
FLT_EPSILON,   DBL_EPSILON,   LDBL_EPSILON	The smallest fraction expressible.