How the answer is -0?

Question

Please forget me if I my question is not right.I am a beginner in c++. I just wrote a simple program in c++ My code is:

#include<iostream>
using namespace std;
int main()
{
    float a,b,c;
    a=0;
    b=-5;
    c=a/b;
    cout<<c;

}

When I ran this code the output is -0. How the answer could be -0 while -0 doesn't exist?

In floating point numbers, there is a single bit for indicating the sign of the number. Depending on how you do the calculation, that can be set, even when the answer comes out as `0`. — BoBTFish, Sep 01 '20 at 10:03
not a duplicate question, but one example for -0 does actually exist: https://stackoverflow.com/questions/45795397/behaviour-of-negative-zero-0-0-in-comparison-with-positive-zero-0-0/45795465 — 463035818_is_not_an_ai, Sep 01 '20 at 10:10
One thing you learn as a programmer is that computer representations of numbers only approximate mathematical notions of what a number is. In computing `-0.0` does exist. — john, Sep 01 '20 at 10:29
@john, I guess you're talking about floating numbers. As on the other number types like int, short,... there is no approximation. — Just Shadow, Sep 01 '20 at 10:31
@JustShadow I'm certainly talking about floating point numbers, but -0.0 is not an approximation for anything, it's a number in it's own right which is different from +0.0. — john, Sep 01 '20 at 10:37
Samsil Arefeen `+0.0 == -0.0` is true, but sometimes +0.0 and -0.0 makes a [difference](https://stackoverflow.com/q/25332133/2410359) . — chux - Reinstate Monica, Sep 01 '20 at 15:13

eerorika · Accepted Answer · 2020-09-01T10:23:04.943

6

How the answer could be -0 while -0 doesn't exist?

Your assumption is wrong. -0 does exist in floating point number representations, and therefore the answer can be -0.

For reference, see the IEEE-754 standard and for "why", see Branch Cuts for Complex Elementary Functions or Much Ado About Nothing's Sign Bit.

edited Sep 01 '20 at 10:23

answered Sep 01 '20 at 10:15

eerorika

232,697
12
197
326

Just Shadow · Answer 2 · 2021-03-18T07:56:13.687

There is a simple logic behind:
Let's assume you're dividing another number to -5, for example, 2.
So if you divide 2 to -5, first it will divide 2 to 5, and then apply the - sign to the result.
The same logic when you use 0 instead of 2. So 0/5 = 0, and applying the - sign, mades it -0.

P.S. As other guys already mentioned in comments, -0 does exist in float universe.
P.P.S. By saying "apply - sign" I mean it just sets the first bit (aka "Sign bit") in the float to 1.

For more info about the float structure see here.

How the answer is -0?

2 Answers2