integer range

[Madan, Gopal]

class Test {
public static void main(String[] args) {
int i = 1000000;
System.out.println(i * i);
}
}

can somebody explain why i am getting -727379968 as the product?
Thanks
Madan

[Daniel Dunleavy]

integer range -2147483648 2147483647

[Madan, Gopal]

Hi Daniel:
I know the integer range (-2,147,483,648 to 2,147,483,647) cannot accomodate product of 1 million by 1 million (1000,000,000,000).
My question is how I got -727,379,968

[Nain Hwu]

My question is how I got -727,379,968

From JLS �15.17.1:

If an integer multiplication overflows, then the result is the low-order bits of the mathematical product as
represented in some sufficiently large two's-complement format. As a result, if overflow occurs, then the
sign of the result may not be the same as the sign of the mathematical product of the two operand values.

1000000*1000000 results in an overflow. The lower-order
32 bits (int is 32 bits) are 0xd4a51000 which is -727,379,968.
[This message has been edited by Nain Hwu (edited October 09, 2001).]

[Madan, Gopal]

Hi Nain:
When i fed the Hex # you said is the low-order 32 bits in NT calc. I am getting 3567587328 and not the number you said/I asked. Also, can you explain/write down how you got 0xD4A51000.
Thanks

[Nain Hwu]

Madan,
I used the toHexString() method in class Integer.
Here is what I did:

lass Test {
      public static void main(String[] args) {
         int i = 1000000;
         int j = i * i;
         System.out.println(Integer.toHexString(j));
         System.out.println(i * i);
      }
}

Here is the output:

D:\pfe101i>java Test
d4a51000
-727379968

Notice that 0xd4a51000 is a negative integer value (in java,
integer is stored as two complement value).
Looks like NT calc treats it as a positive value, which gives
you 3567587328.
Hope this help.