Where am I wrong??

[sonir shah]

    public class Test017
    {    
        public static void main(String args[])
        {
           System.out.println(Math.abs(Byte.MIN_VALUE) > 0);
           System.out.println(Math.abs(Long.MIN_VALUE) > 0);
           System.out.println(Math.abs(Short.MIN_VALUE) > 0);
           System.out.println(Math.abs(Float.MIN_VALUE) > 0);
        }
    }

Options:
A: true true true true
B: true false true true
C: true true true false
D: false false false true
Answer: b)true false true true
I tried running the program and the output was b).
I feel that min value of byte is -127 which is less than 0 hence false should be the first part..correct me
I am not able to understand how can we get such an output.
Please explain
Sonir

[Shivaji Marathe]

Sonir :
What does Math.abs() do ?

[sonir shah]

Returns an Absolute value.
Sonir

[mark stone]

it gives you the abosolute value of a number
like for -5 abs value would be 5
for -.0075 abs value would be .007
and so on....

Originally posted by Shivaji Marathe:
Sonir :
What does Math.abs() do ?

[Shivaji Marathe]

Sonir :
If you know what math.abs() returns , then you know that abs() of any negative number is it's positive value - do you not?

Originally posted by sonir shah:
Returns an Absolute value.
Sonir

[Shivaji Marathe]

Actually I am surprised that math.abs(Long.min_value ) prints a negative number. Is this a normal behavior or is this an " un documented feature?"

[mark stone]

they should all be true ? right ?
why is Long.min_value giving false ?

Originally posted by Shivaji Marathe:
Actually I am surprised that math.abs(Long.min_value ) prints a negative number. Is this a normal behavior or is this an " un documented feature?"

[Maulin Vasavada]

hi sonir,
the problem here is this.
as per API says, when the value is MIN_VALUE and the datatype is int or long then it crosses the limit if we make it +ve as there is one number less in +ve than in +ve. so, when we do that it wraps the number again and make it -ve.
thus, comparision to >0 returns false.
for byte & short there is no method in Math.abs() list. so short and byte are first converted to int and then it works. here Byte.MIN_VALUE is -128 but still it works as it is fully representable in int. same for short. and so making them abs() returns int > 0.
however, if we had Integer.MIN_VALUE or Long.MIN_VALUE then it would return the same number which is -ve due to the reason explained above.
this reasoning doesnt apply to float and double. as the Float.MIN_VALUE is +/-1.4E-45 and similar thing for double. so, when u apply abs() to Float.MIN_VALUE or Double.MIN_VALUE it gives u the +ve number and >0 comparison is true.
so, u get results u r having.
regards
maulin.

[mark stone]

mauline i don;t get this why all are not true...

Originally posted by Maulin, Vasavada:
hi sonir,
the problem here is this.
as per API says, when the value is MIN_VALUE and the datatype is int or long then it crosses the limit if we make it +ve as there is one number less in +ve than in +ve. so, when we do that it wraps the number again and make it -ve.
thus, comparision to >0 returns false.
for byte & short there is no method in Math.abs() list. so short and byte are first converted to int and then it works. here Byte.MIN_VALUE is -128 but still it works as it is fully representable in int. same for short. and so making them abs() returns int > 0.
however, if we had Integer.MIN_VALUE or Long.MIN_VALUE then it would return the same number which is -ve due to the reason explained above.
this reasoning doesnt apply to float and double. as the Float.MIN_VALUE is +/-1.4E-45 and similar thing for double. so, when u apply abs() to Float.MIN_VALUE or Double.MIN_VALUE it gives u the +ve number and >0 comparison is true.
so, u get results u r having.
regards
maulin.

[sonir shah]

Maulin,
Even I am not clear why is it giving false for the second one..
Can you give me some examples??
Sonir

[Maulin Vasavada]

hi sonir,
say u have 4-bit capacity registers. so we can have -2^3-1 to +2^3 numbers that is [-8,7] (keep it int tho).
why so? how do v decide -ve limit?
simple. comp uses 2's complement. so startout writing +ve numbers from 0 along with it's -ve equvivalent.
0 = 0000 , -0 = 0000 (which isnt really there. so we will discard -0)
1 = 0001 , -1 = 1111
2 = 0010 , -2 = 1110
3 = 0011 , -3 = 1101
4 = 0100 , -4 = 1100
5 = 0101 , -5 = 1011
6 = 0110 , -6 = 1010
7 = 0111 , -7 = 1001
so far as u can see all the -ve have 1 as Most Significant Bit (MSB). so that should mean that when we try to represent +ve 8 = 1000 it is not +ve but -ve.
so, we have +7 max and -8 min but dont have +8.
now, when we take abs() of any -ve value see what happens,
e.g. -3 = 1101. we can just do that by taking 2's complement which would be 0011 = 3. right?
now take -8 = 1000. get 2's complement. result is 1000 = -8 !!!
so, MIN_VALUE will result into the same value for int and long for this reason. and so if we do comparison with 0 it will be false for (number > 0).
hope u got this.
regards
maulin.

[Shivaji Marathe]

Thanks Maulin. I get it. very tricky Q.