Bitwise inversion operator question

[jesse]

public class test{
public static void main(String args[]){
byte x=3;
x=(byte)~x;
System.out.println(x);
}
}
A. 3
B. 0
C. 1
D. 11
E. 252
F. 214
G. 124
H. -4
The correct answer was given as H. The answer i got was -124. Could someone please explain this question to me.

[Paul Anilprem]

I'm not sure how are you getting -124. Did you type it exactly as you've given here?
-Paul.
------------------
Get Certified, Guaranteed!
(Now Revised for the new Pattern)
www.enthuware.com/jqplus

[Rob Levo]

~n = -n - 1
Can anyone confirm this?

[Paul Anilprem]

That's correct. ( but only if there is no overflow etc.) For eg. It won't work in cases like this:
int x = 128;
byte b = (byte) ~x; //here b will be 127! Now figure it out.
------------------
Get Certified, Guaranteed!
(Now Revised for the new Pattern)
www.enthuware.com/jqplus

[sola yemi]

public class test{
public static void main(String args[]){
byte x=3;
x=(byte)~x;
System.out.println(x);
}
}
the ~ operator simply works like this.
~X is the same as (-X)-1: //just change the sign and deduct 1.
//in this case
~x = (-3)-1
which is the same as -4.

[sola yemi]

Thanks for the info about the overflow, if i'd seen this in the exam, i would have been totally confused, thanks again.

Originally posted by Paul Anil:
That's correct. ( but only if there is no overflow etc.) For eg. It won't work in cases like this:
int x = 128;
byte b = (byte) ~x; //here b will be 127! Now figure it out.

[Dev Prakash]

3 in binary(byte) is 00000011
if you inverse it, it should become 11111100 which is equal to 252 in decimal.
Is that not correct???
Please explain...

[srival vas]

3 in binary(byte) is 00000011
if you inverse it, it should become 11111100 which is equal to 252 in decimal.
Is that not correct???
Please explain...
------------
Dev, Signed bit and 2's complement , that is the catch.

[srival vas]

quote:
--------------------------------------------------------------------------------
Originally posted by Paul Anil:
That's correct. ( but only if there is no overflow etc.) For eg. It won't work in cases like this:
int x = 128;
byte b = (byte) ~x; //here b will be 127! Now figure it out.
-----------------------------------------------------------------
just to clarify though I am sure you all realised it,
b is 127 not bcos of ~, but bcos -129 (-128 -1 ) has
been casted to byte, I would assume I can still safely
assume ~n = -n -1 and then concern myself with the casts et al.

[psr krishna]

hi jesse,
The answer for your question is -4. how?
here is explanation....
public class test{
public static void main(String args[]){
byte x=3;
x=(byte)~x;
System.out.println(x);
}
}
A. 3
B. 0
C. 1
D. 11
E. 252
F. 214
G. 124
H. -4

byte x=3 is in 8 bit representation so, in binary it is 00000011 OK! then
~x= bit wise inversion is 11111100 OK! after this inversion
when right most bit is 1 it will be considered as a
negative value and it is assigned to x.In JVM the negative numbers will be represented in 2's complement form ,so to get actual number you first again inverse it and add 1 to last significant bit and express your result in nagative value, so this is like this
x: 00000011
~x: 11111100

for result inverse the above ~(~x): 00000011
add 1 to left most bit 1
--------
00000100 which is 4 in
--------
decimal notation and express result in negative i.e.:-4
OK! NOW I THINK YOUR DOUBT IS CLEARED.
---THANKS
SHIVARAM

[psr krishna]

hi jesse,
please correct a little mistake in my previous post

byte x=3 is in 8 bit representation so, in binary it is 00000011 OK! then
~x= bit wise inversion is 11111100 OK! after this inversion
when left most bit is 1 it will be considered as a
negative value and it is assigned to x.In JVM the negative numbers will be represented in 2's complement form ,so to get actual number you first again inverse it and add 1 to right significant bit and express your result in nagative value, so this is like this
x: 00000011
~x: 11111100

for result inverse the above ~(~x): 00000011
add 1 to right most bit 1
--------
00000100 which is 4 in
--------
decimal notation and express result in negative i.e.:-4
OK! NOW I THINK YOUR DOUBT IS CLEARED.
---THANKS
SHIVARAM

[Raghvendra Sharma]

The answer is -4 only.
Try it with the jdk and see the result yourself.
3 decimal = 0000 0011 ( binary )
when you invert it becomes 1111 1100 ( binary )
That's 2complement representation of -4 ( check by 1. inverting and 2. adding 1 to result of inversion ).
You should check once more.
i hope it helps.
raghav..