MORE ABOUT FLOAT AND DOUBLE VARIABLES
C displays both float and double variables to six decimal places. This does NOT refer to the precision (accuracy) of which the number is actually stored, only how many decimal places printf() uses to display these variable types.
The following program illustrates how the different data types are declared and displayed,

#include <stdio.h>

main()
{
int sum = 100;
char letter = 'Z';
float set1 = 23.567;
double num2 = 11e+23;

printf("Integer variable is %d\n", sum);
printf("Character is %c\n", letter);
printf("Float variable is %f\n", set1);
printf("Double variable is %e\n", num2);
}

Sample program output
Integer variable is 100
Character variable is Z
Float variable is 23.567000
Double variable is 11.000000e23

To change the number of decimal places printed out for float or double variables, modify the %f or %e to include a precision value, eg,

printf("Float variable is %.2f\n", set1 );
In this case, the use of %.2f limits the output to two decimal places, and the output now looks like
Sample program output
Integer variable is 100
Character variable is Z
Float variable is 23.56
Double variable is 11.000000e23

 

SPECIAL NOTE ABOUT DATA TYPE CONVERSION
Consider the following program,

#include <stdio.h>

main()
{
int value1 = 12, value2 = 5;
float answer = 0;

answer = value1 / value2;
printf("The value of %d divided by %d is %f\n",value1,value2,answer );
}

Sample program output
The value of 12 divided by 5 is 2.000000
Even though the above declaration seems to work, it is not always 100% reliable. Note how answer does not contain a proper fractional part (ie, all zero's).
To ensure that the correct result always occurs, the data type of value1 and value2 should be converted to a float type before assigning to the float variable answer. The following change illustrates how this can be done,

answer = (float)value1 / (float)value2;

DIFFERENT TYPES OF INTEGERS
A normal integer is limited in range to +-32767. This value differs from computer to computer. It is possible in C to specify that an integer be stored in four memory locations instead of the normal two. This increases the effective range and allows very large integers to be stored. The way in which this is done is as follows,

long int big_number = 245032L;
To display a long integer, use %l, ie,

printf("A larger number is %l\n", big_number );
Short integers are also available, eg,

short int small_value = 114h;
printf("The value is %h\n", small_value);
Unsigned integers (positive values only) can also be defined.

The size occupied by integers varies upon the machine hardware. ANSI C (American National Standards Institute) has tried to standardise upon the size of data types, and hence the number range of each type.

The following information is from the on-line help of the Turbo C compiler,
Type: int
Integer data type

Variables of type int are one word in length.
They can be signed (default) or unsigned,
which means they have a range of -32768 to
32767 and 0 to 65535, respectively.

Type modifiers: signed, unsigned, short, long

A type modifier alters the meaning of the base
type to yield a new type. Each of the above
can be applied to the base type int. The
modifiers signed and unsigned can be applied
to the base type char. In addition, long can
be applied to double. When the base type is
ommitted from a declaration, int is assumed.

Examples:
long x; /* int is implied */
unsigned char ch;
signed int i; /* signed is default */
unsigned long int l; /* int ok, not needed */