[SOLVED] Bit "overflow" for negative numbers

atlantis43 · 03-20-2015, 11:31 PM

This may be a trivial question, but I'm wondering if there is any way of expressing an overflow bit for a negative value while performing Bit manipulations in C Programming (if there would ever be such a need)?
Is there any equivalent of (1)00001011 for a negative?
Thanks

mdooligan · 03-20-2015, 11:35 PM

That's the high bit. Bit 8. 1000 0001 is -1 (signed char).

NevemTeve · 03-21-2015, 12:35 AM

What bit manipulation do you mean? Shift (<< >>)?

ntubski · 03-21-2015, 10:12 AM

Quote:

Originally Posted by atlantis43

I'm wondering if there is any way of expressing an overflow bit for a negative value

I don't understand what you mean by "expressing an overflow bit".

Quote:

Originally Posted by mdooligan

1000 0001 is -1 (signed char).

Not if the machine uses 2's complement, which is the case for pretty much all computers in use today.

If I recall correctly the C standard doesn't specify which encoding to use for signed numbers.

atlantis43 · 03-21-2015, 10:20 AM

Quote:

Originally Posted by NevemTeve

What bit manipulation do you mean? Shift (<< >>)?

I guess I'm referring more to simple problems like bit-wise addition % subtraction, but I think that I've "seen the light" that my question is not relevant.
Thanks anyhow.

mdooligan · 03-22-2015, 11:03 AM

Quote:

Originally Posted by ntubski

Not if the machine uses 2's complement, which is the case for pretty much all computers in use today.

Oops. My bad. That was a quickie post before I signed off for the night. I must have been sleepy

Code:

1111 1111 = -1, 
1000 0001 = -127, 
1000 0000 = -128.
^
High bit.