BASIC ELEMENTS OF JAVA - Java Programming: From Problem Analysis to Program Design - page 35

To represent real numbers, Java uses a form of scientific notation called floating-point
notation. Table 2-3 shows how Java might print a set of real numbers. In the Java
floating-point notation, the letter e stands for the exponent.

2

TABLE 2-3 Examples of Real Numbers Printed in Java Floating-Point Notation

Real Number Java Floating-Point Notation

75.924 7.592400e+01

0.18 1.800000e-01

0.0000453 4.530000e-05

-1.482 -1.482000e+00

7800.0 7.800000e+03

Java provides two data types to represent decimal numbers: float and double . As in the
case of integral data types, the data types float and double differ in the set of values.
float : The data type float is used in Java to represent any real number between
3.4E+38 and 3.4E+38. The memory allocated for the float data type is 4 bytes.
double : The data type double is used in Java to represent any real number between
1.7E+308 and 1.7E+308. The memory allocated for the double data type is 8 bytes.
Other than the set of values, there is one more difference between the data types float
and double . The maximum number of significant digits—that is, the number of decimal
places—in float values is 6 or 7. The maximum number of significant digits in values
belonging to the double type is typically 15. The maximum number of significant digits
is called the precision. Sometimes float values are called single precision, and values
of type double are called double precision.

In Java, by default, floating-point numbers are considered to be of type double .
Therefore, if you use the data type float to represent floating-point numbers in a
program, you might get a warning or an error message, such as ''truncation from
double to float'' or ''possible loss of data.'' To avoid such messages, you should use
the double data type. For illustration purposes and to avoid such messages in
programming examples, this topic mostly uses the data type double to represent
floating-point numbers. However, if you need to explicitly use a float value in a
program, specify that the decimal value is a float value using the letter f at the
end of the number. For example, 28.75f represents a float value while 28.75
represents a double value.