What is the meaning of `-1.$e+000`?

Question

What is the meaning of the symbol -1.$e+000? In what category are these kind of things included or with what keywords I can search them?

enter image description here

EDIT: After increasing precision this is what I got:

enter image description here

It's not really a "symbol" per se. I suppose that depends on what abstraction layer you're talking about, though. :) — Lightness Races in Orbit, Jan 31 '12 at 14:47
Not without knowing which direction you want me to be more explicit in. What do you want to know? — Lightness Races in Orbit, Jan 31 '12 at 14:56
Related: http://stackoverflow.com/questions/840081/what-does-floating-point-error-1-j-mean — Lightness Races in Orbit, Jan 31 '12 at 15:00
I didn't want a screenshot. You might as well have scribbled it on a piece of paper and taken a photograph. >. — Lightness Races in Orbit, Jan 31 '12 at 15:06

score 6 · Answer 1 · edited Jan 31 '12 at 14:47

6

edited Jan 31 '12 at 14:47

answered Jan 31 '12 at 14:46

mskfisher

2

The original question asked about `-1.#INF` and, after I answered, was edited to `-1.$e+000`, which is not mentioned in the article I linked. I'm not sure they are the same. – mskfisher Jan 31 '12 at 14:51
1

[Looks like NaN from Windows, with precision set small and a scientific exponent suffix.](http://stackoverflow.com/a/5542066/560648) – Lightness Races in Orbit Jan 31 '12 at 15:02
The edit history doesn't confirm that, but somehow I believe it's true. – Daniel Fischer Jan 31 '12 at 15:03
@Daniel: It might have been edited within the five-minute grace period. – Lightness Races in Orbit Jan 31 '12 at 15:05
The edit history doesn't show edits made within the first 5 minutes: http://meta.stackexchange.com/questions/44174/edited-answer-does-not-show-edited-or-history-link – mskfisher Jan 31 '12 at 15:06
That's why I didn't disbelieve it. – Daniel Fischer Jan 31 '12 at 15:08
@DanielFischer: Nondisbeliever – Lightness Races in Orbit Jan 31 '12 at 15:19
@Lightness I don't believe that. – Daniel Fischer Jan 31 '12 at 15:50
@Daniel: I hope you're not dissing – Lightness Races in Orbit Jan 31 '12 at 15:50

score 5 · Accepted Answer · edited May 23 '17 at 10:32

5

When you see -1.#INF, -1.#IND and similar text for formatted numerical output, you're on Microsoft Visual Studio and you're seeing a special floating-point type that cannot be represented by a number... that is, either infinity or NaN ("not a number").

The e+000 is just a suffix from scientific notation, indicating an exponent of 0. It's safely ignored.

On Linux GCC you tend to see the text inf and nan instead.

edited May 23 '17 at 10:32

Community

answered Jan 31 '12 at 15:05

@mskfisher: I certainly didn't intend for this answer to overtake yours.. It was supposed to be an addendum! Sorry ;( – Lightness Races in Orbit Jan 31 '12 at 15:19

2 Answers2