Why (int)(14/13 - 0.001) yield 0 and not 1?

Question

Isn't it true that if you cast a float number like 1.0012 to an integer, it will become 1?

Then why is it when I write:

(int)(14/13-0.001)

instead of 1.07592 ~ become 1 it becomes 0 ?

(Java compiled with Eclipse).

[What Every Computer Scientist Should Know About Floating-Point Arithmetic](http://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html) — phuclv, Mar 02 '14 at 08:51
the title of question is not correct, `(int)(1.01 - 0.01)` will produce 1 not 0. — sakura, Mar 02 '14 at 08:55
14/13 is equal to 1 if they are integers - and they are in your example. Subtracting .01 and then taking the it will produce 0 — Rob, Mar 02 '14 at 08:55
[is floating point math broken](http://stackoverflow.com/questions/588004/is-floating-point-math-broken) — Lutz Lehmann, Mar 02 '14 at 08:56
In primary school you might have done integer division where something like `14 divided by` 13 is `1 with 1 remainder`. This is how integer arithmetic works and you can do `14 % 13` to get the remainder. — Peter Lawrey, Mar 02 '14 at 09:06
@LưuVĩnhPhúc Thanks for your suggestion, but if you actually take the time to read the question, you will see that the OP is not at the stage where the document you provide a link to is likely to help. — Pascal Cuoq, Mar 02 '14 at 17:51

Dejan · Accepted Answer · 2014-03-02T09:03:19.023

12

It does truncating. For 1.0012 it just removes part on right of decimal point.

In example

(int)(14/13-0.001)

14/13 will be 1 and then it will be converted to double, 1.0-0.001 = 0.999, and after truncating it becomes 0.

edited Mar 02 '14 at 09:03

answered Mar 02 '14 at 08:52

Dejan

1

It will be converted to double, not float. – JB Nizet Mar 02 '14 at 08:55
If I remember correctly it will be float, but the result will be same nonetheless. – Dejan Mar 02 '14 at 08:57
http://docs.oracle.com/javase/tutorial/java/nutsandbolts/datatypes.html: *A floating-point literal is of type float if it ends with the letter F or f; otherwise its type is double and it can optionally end with the letter D or d.* – JB Nizet Mar 02 '14 at 08:58
Yep, I think you are right, I checked documentation. – Dejan Mar 02 '14 at 09:02
1

it's really count 14/13 as interger so i had to write it as (float)14/13-.001 – Ricky Leo Saputra Mar 02 '14 at 09:04
14.0/13 or 14/13.0 is enough, shorter and easier to type, no need to cast – phuclv Mar 04 '14 at 14:10

score 8 · Answer 2 · answered Mar 02 '14 at 08:52

8

14 / 13 is an integer division. Its value is 1.

1 - 0.01 is thus lower than 1.

Casting it to int thus yields 0.

Use 14.0 / 13 - 0.001 or 14d / 13 - 0.001 instead.

answered Mar 02 '14 at 08:52

JB Nizet

2 Answers2