Javascript substracting numbers is not accurate - why

Question

I have a fiddle here: http://jsfiddle.net/94wQJ/1/ - but probably someone can advise just by looking below.

<button type="button" id="allocate">Calc</button>

 $('#allocate').click(function () {
     val1 = 25.00;
     val2 = 16.37;
     val3 = val1-val2;
     alert(val3);
 });

25 - 16.37 = 8.63 - however, the alert for val3 = 8.62999999999

Why is it not accurate?

Thank you,

Mark

some good info here http://stackoverflow.com/questions/588004/is-javascripts-floating-point-math-broken — andy mccullough, Sep 27 '13 at 08:31
elaborate discussion here http://stackoverflow.com/questions/11832914/round-up-to-2-decimal-places-in-javascript and here http://stackoverflow.com/questions/7342957/how-do-you-round-to-1-decimal-place-in-javascript — hima, Sep 27 '13 at 08:38

score 5 · Accepted Answer · edited May 23 '17 at 12:34

5

Try Using toFixed more information here

val3.toFixed(2)

Demo Fiddle

edited May 23 '17 at 12:34

Community

1
1

answered Sep 27 '13 at 08:30

Greenhorn

1,690
11
12

Thank you Ram, I'll mark as the answer as soon as SO allows. Strange behaviour!! Cheers, Mark – Mark Sep 27 '13 at 08:31
1

@MarkTait - It is not strange behaviour - It is due to the limitations (known) of floating point numbers – Ed Heal Sep 27 '13 at 08:51

score 3 · Answer 2 · edited May 23 '17 at 10:10

3

Why is it not accurate?

This is a duplicate of Is floating point math broken?, but to answer your specific question: floating point numbers generally store the number in base 2 because it allows storing more numbers more accurately than using base 10, at the expense of not being able to exactly store all base 10 numbers even with a small number of decimal places.

edited May 23 '17 at 10:10

Community

1
1

answered Sep 27 '13 at 08:41

Douglas

36,802
9
76
89

its not just Javascript's floating point math - it's floating point math in general unless you use things like specialised Decimal classes. – sevenseacat Sep 27 '13 at 09:08
@sevenseacat: yes, agreed. You could go further and say that it is a limitation on how much information can you store in a particular number of bits, I can't quote which bit of information theory would apply though. – Douglas Sep 27 '13 at 09:11

PSR · Answer 3 · 2013-09-27T08:32:33.040

1

Many decimal fractions not be represented exactly in binary.

Use .toFixed(2)

edited Sep 27 '13 at 08:32

answered Sep 27 '13 at 08:27

PSR

39,804
41
111
151

score 1 · Answer 4 · answered Sep 27 '13 at 08:35

1

Please check this

val3 = Math.round(val3*100)/100;

I have update here http://jsfiddle.net/94wQJ/7/

answered Sep 27 '13 at 08:35

hima

610
3
10
24

Javascript substracting numbers is not accurate - why

4 Answers4