Clarify the output please . . .

[Anonymous]

Guys, Can you please tell me why I get this output from the code below.
Yikes. a and b are equal
Nope. a and c are not equal
Appreciate your help.
----------------------------------------------------------------
import java.math.*;
public class Q1 {
public static void main(String[] args)
{
double a = Math.ceil(Double.MIN_VALUE);
double b = Math.ceil(Double.MIN_VALUE + 1);
double c = Math.ceil(Double.MIN_VALUE + 2);
if (a == b)
System.out.println("Yikes. a and b are equal");
else
System.out.println("Nope. a and b are not equal");
if (a == c)
System.out.println("Yikes. a and c are equal");
else
System.out.println("Nope. a and c are not equal");
}
}

[Anonymous]

Double.MIN_VALUE is the smallest positive value a double variable can hold, which is 4.9E-324.
( I think, 4.9E-324 = (4.9/(Math.pow(10,324))
Math.ceil(Double.MIN_VALUE) is equal to 1.0. [because ceil returns the smallest double value that is not less than the argument and is equal to a mathematical integer.]
Math.ceil(Double.MIN_VALUE + 1) is equal to 1.0, because if you add Double.MIN_VALUE to any variable, it does not affect the value of that variable.(May be because Double.MIN_VALUE is so negligible. I dont know why, but this is what it does).
double d=1.0;
d=d+Double.MIN_VALUE;
System.out.println(d); // prints 1.0, not 1.0 + 4.9E-324
That is why
Math.ceil(Double.MIN_VALUE)==Math.ceil(Double.MIN_VALUE + 1);
is true.
Because both operands evaluate to be 1.0
- Srini

[Anonymous]

Thanks Srini for the explanation.
I understand it better now, though I fail to understand why Math.ceil(some_integer + Double.MIN_VALUE (which is another positive value)) is still some_integer.
But I will just remember this point for the exam.
Thanks once again !!!
Regards,
Shashank

[rajsim]

Any value less than 1e-16 is being ignored. Is it a bug?