# Why fprintf('%​3.1f',roun​d(-eps)) prints -0.0?

16 views (last 30 days)
Sandor Toth on 17 Jan 2017
Commented: James Tursa on 17 Jan 2017
I found some very strange behaviour of the round function in Matlab R2016b (at least for me). I would expect that the round function returns 0 for -eps, but surprisingly the command:
num2hex(round(-eps))
Returns:
8000000000000000
num2hex(0) == 0000000000000000
This causes fprintf() to print -0.0. Is this a bug or I just miss something?
James Tursa on 17 Jan 2017
Side Note: This seems to be consistent among the related functions ceil and fix as well. E.g.,
>> num2hex(round(-eps))
ans =
8000000000000000
>> num2hex(ceil(-eps))
ans =
8000000000000000
>> num2hex(fix(-eps))
ans =
8000000000000000

Stephen on 17 Jan 2017
Edited: Stephen on 17 Jan 2017
You are missing the fact that MATLAB has two zeros: positive zero and negative zero, which have different HEX values. This is a result of the IEEE 754 floating point standard that MATLAB implements, which uses one bit to indicate the number sign: a corollary of this is that even a zero can be positive or negative.
It is not possible to use < to directly test if zero is negative (by definition negative zero is equal to positive zero, so this makes sense):
>> (-0)<0
ans =
0
but here is one simple trick that will test for negative zero:
>> (1/-0)<0
ans =
1
Which relies on this behavior:
>> 1/-0
ans =
-Inf
ans =
-Inf
Neat huh!
You can find more interesting information on negative zeros here:

Edited: Adam on 17 Jan 2017