7.1 Ap0010: self-assessment, primitive types  (Page 8/9)

What is the value range for a float?

In case you are interested, using the same syntax as above, the value range for type float is from 1.4E-45 to 3.4028235E38

Double is often the default type

There is another thing that is significant about type double . In many cases where a value is automatically converted to a floating type, it isconverted to type double rather than to type float. This will come up in future modules.

Back to Question 4

Answer 3 .

A. -2147483648

Explanation 3

Could easily have guessed

As a practical matter, you had one chance in two of guessing the correct answer to this question, already having been given the value of the largestalgebraic value for type int.

And the winner is ...

Did you answer B. -2147483647? -- WRONG

If so, you may be wondering why the most negative value isn't equal to thenegative version of the most positive value?

A twos-complement characteristic

Without going into the details of why, it is a well-known characteristic of binary twos-complement notation that the value range extends one unit further inthe negative direction than in the positive direction.

What about the other two values?

Do the values of -32768 and 32767 in the set of multiple-choice answers to this question represent anything in particular?

Yes, they represent the extreme ends of the value range for a 16-bit binary number in twos-complement notation.

Does Java have a 16-bit integer type?

Just in case you are interested, the short type in Java is represented in 16-bit binary twos-complement signed notation, so this is thevalue range for type short.

What about type byte?

Similarly, a value of type byte is represented in 8-bit binary twos-complement signed notation, with a value range extending from -128to 127.

Back to Question 3

Answer 2 .

B. 2147483647

Explanation 2

First question on types

This is the first question on Java types in this group of self-assessmentmodules.

32-bit signed twos-complement integers

In Java, values of type int are stored as 32-bit signed integers in twos-complement notation.

Can you calculate the values?

There are no unsigned integer types in Java, as there are in C++. If you are handy with binary notation, you could calculate the largest positivevalue that can be stored in 32 bits in twos-complement notation.

See documentation for the Integer class

Otherwise, you can visit the documentation for the Integer class, which provides a symbolic constant (public static final variable) named MAX_VALUE . The description of MAX_VALUE reads as follows:

"The largest value of type int. The constant value of this field is 2147483647."

Back to Question 2

Answer 1 .

C. Hello World

Explanation 1

The answer to this first question is intended to be easy. The purpose of the first question is to introduce you to the syntax that will frequently be usedfor program code in this group of self-assessment modules.

The controlling class and the main method

In this example, the class named Ap001 is the controlling class . It contains a method named main , with a signature that matches the required signature for the main method. When the user executes this program, the Java virtual machine automatically calls the method named main in the controlling class.