C Float Types

Summary: in this tutorial, you will learn about various C float types including float, double and long double.

Introduction to C float types

C provides various floating-point types that represents non-integer number with a decimal point at any position. For example, with integer types, you only can have numbers 1, 2, 10, 200… however with floating-point type, you can have 1.0, 2.5, 100.25 and so on.

There are three standard floating-point types in C:

• float: for numbers with single precision.
• double: for numbers with double precision.
• long double: for numbers with extended precision.

The following table illustrates the technical attributes of various floating-point types in C. It is important to notice that this is only the minimal requirement for storage size defined by C.

What is precision?

If you consider the fraction 5/3, this number can be represented in decimal as 1.666666666... with infinite 6. When you use the floating-point type with limited bytes to store infinite number, it is impossible. Therefore the floating-point value can be stored only with a limited precision. The precision is expressed as a number of significant digits or in another way, the precision is defined as how many digits that a number can represent without data loss.

For example, the float type with 4 bytes = 4 x 8 bits = 32 bits, is for numbers with single precision. It means that a float gives 1 sign bit, 8 bits of exponent and 23 bits of significand.

The double type is for number with double precision that gives 1 sign bit, 11 bits of exponent and 52 bits of significand. (8 bytes x 8 bits = 64 bits = 1 bits + 52 bits + 11 bits).

Float ranges and precision

In order to find the value ranges of floating-point number in your platform, you can use the float.h header file. This header file defines macros such as FLT_MIN, FLT_MAX and FLT_DIG that store the float value ranges and precision of the  float type.

You can also find the corresponding macros for double and long double with the prefixes  DBL_ and LDBL_

The following program illustrates the storage size and precision of floating-point numbers in your system.

/*
 * File   : main.c
 * Author : zentut.com
 * Purpose: C float demo
 */

#include <stdio.h>
#include <float.h>

int main( )
{
  printf("Storage size: %d bytes\n"
         "Minimum float value: %E\n"
         "Maximum float value: %E\n"
         "Precision: %d decimal digits\n",
         sizeof(float),
         FLT_MIN,
         FLT_MAX,
         FLT_DIG);

  puts("\nExample of float precision:\n");
  double d = 12345.6;
  float  f = (float)d;

  printf("The floating-point number d  "
         "%18.10f\n", d);
  printf("Stored in a variable f\n"
         "of type float as the value   "
         "%18.10f\n", f);

  return 0;
}

The output of the program in our system is as follows:

Storage size: 4 bytes
Minimum float value: 1.175494E-038
Maximum float value: 3.402823E+038
Precision: 6 decimal digits

Example of float precision:

The floating-point number d    12345.6000000000
Stored in a variable f
of type float as the value     12345.5996093750

In this tutorial, we have introduced you various C float types and explained what precision means to each type of float in C.