20140 - Rounding Error?

Status: RESOLVED DUPLICATE of bug 5856

(Core :: JavaScript Engine, defect, P3)

Description

[Sitsofe Wheeler]

The following does not produce the expected answer:
<script language="javascript1.2"><!--
document.write("14.28 x 9 should be <b>128.52</b>;<br><small>your browsers'
answer = </small><b>" + (14.28*9)+"</b>");
// --></script>

Expected:
128.52

Result:
128.51999999999998

However, both IE 4.01 and Netscape 4.5 on the PC produce the same error. I
don't think that this makes it right though...

System:
Build ID: 1999111520 WinNT 4 SP 6

Comment 1

[Mike McCabe]

I hate to say this, but it's not an error!

JavaScript (like most programming languages) computes values using floating
point numbers in a binary representation.  Unfortunately, this format can't
exactly represent many numbers we can represent exactly using decimal notation.
One such number is 14.28... So when the decimal numbers we work with in
JavaScript are converted to binary floating point representation internally,
approximations are sometimes introduced; these sometimes show up in the output,
as you noticed.  For better or worse, this behavior is specified by the ECMA-262
standard to which JavaScript conforms.

Thanks for the report!  Sorry I can't give a better answer; if this is a problem
for you, I believe that there are rounding routines available for JavaScript
that should give you reasonable result.

Comment 2

[Waldemar Horwat]

This behavior is covered both by the ISO standard 16262 (ECMAScript) and the

ANSI/IEEE standard 754.  You will find the same behavior in most other

programming languages, including Java and almost all implementations of C/C++.

If our implementation of JavaScript did not produce 128.51999999999998, we would

be violating international floating-point arithmetic standards.



In JS1.5 you can use Number.toFixed to round the answer to the number of decimal

places you desire.

Comment 3

[Keyser Sose]

Mozilla compilies with Standards so its all good :)

Comment 4

[Josh Birnbaum]

*** Bug 202827 has been marked as a duplicate of this bug. ***

Comment 5

[Steven Cole]

*** Bug 210874 has been marked as a duplicate of this bug. ***

Comment 6

[Jo Hermans]

*** Bug 221816 has been marked as a duplicate of this bug. ***

Comment 7

[Chris Thomas (CTho) [formerly [e-mail] [e-mail]]]

*** Bug 261947 has been marked as a duplicate of this bug. ***

Comment 8

[Dan Erikson]

*** Bug 280890 has been marked as a duplicate of this bug. ***

Comment 9

[:Gavin Sharp [email: [e-mail]]]

*** Bug 280890 has been marked as a duplicate of this bug. ***

Comment 10

[Jesse Ruderman]

*** This bug has been marked as a duplicate of 5856 ***