Reference on primitive types - Multimedia Introduction to Programming Using Java - page 221

// An example of inexact results and rounding:
System.out.print("inexact results with float: ");
for ( int i= 0; i < 1000; i= i + 1) {
double z= 1.0D / i;
if (z * i != 1.0D) {
System.out.print( " " + i);

}

}
A complete discussion of double numbers is beyond the scope of this topic.
For more information about double , look at the specification of Java and turn to
more advanced topics on programming and numerical analysis.

6.4.2 Type float

Type float is similar to type double , except that a float value occupies four
bytes instead of eight. Its negative and positive values are:

-3.4028235E38 to -1.4E-45
1.4E-45 to 3.4028235E38
In most situations, type double is used and not float because the use of
double gives much more accuracy. However, when space is at a premium and
accuracy is not important, use float .

float literals
To change a double literal (that does not include d or D ) into a float literal,
follow it (with no separating whitespace) by a lower case or upper case F . For
example, 3.0 and 5D are double literals, but 3.0F and 5F are float literals.

Operations of type float
The operations of type float are similar to those of type double . They are:
negation, unary addition, addition, subtraction, multiplication, division, and
remainder. Given float operands, they produce a float result. See Sec. 6.1 for
a description of the operations and Sec. 6.6 for a discussion of casting.

6.5 Type char

Type char has as its values the set of characters that you can process in a Java
program.
The constants of this type, called literals of type char , are single characters
enclosed in single-quote marks, e.g. 'A' and ';' . A sequence of these characters
enclosed within double quotes " form a literal of class String . For example,
String literal

Activity
13-7

"A;a"