When converting from a float value to an integer, if the float value is too 
large, the result is 0 instead of the maximum integer. Here's the test.
class FloatBug {
    public static void main ( String[] args ) {
        int i = (int) 3.4e+32f;
        System.out.println ( "(int) 3.4e+32f = " + i );
    }
}
The answer should be 2147483647.