Re: Simple problem for float calculation

Article View: borland.public.jbuilder.java.api

Article #2537

Back to borland.public.jbuilder.java.api Previous Next

#2537

From: Joi Ellis
Date: Tue, 17 Oct 2000 00:00

29 lines
1452 bytes

eLe <petruschan@realink.com.hk> wrote:
> I have Java program simple as :

> float a = 27.1
> float b = 0.2
> float c = a + b
> c = 27.300001

> I want the ans is c = 27.3 is not 27.300001

> How to solve this problem ? thank you !

Decimals like 0.1 are irrational numbers in binary.  You will see
round off errors like this, they are unavoiable.

If you're doing math with money, express the money values in pennies, using
integers variables.

Otherwise, you must change your expression to round the result to the desired
number of significant digits yourself.

-- 
Joi Ellis                    Software Engineer
Aravox Technologies          joi@aravox.com, gyles19@visi.com

No matter what we think of Linux versus FreeBSD, etc., the one thing I
really like about Linux is that it has Microsoft worried.  Anything
that kicks a monopoly in the pants has got to be good for something.
           - Chris Johnson

Technical Details

Message-ID: <8shoqj$n2q16@bornews.borland.com>

Path: arc1-ca.pugleaf.net!unknown!not-for-mail

References: <39ec3576_2@dnews>

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 1089d0,9bcc96b038465cb3
X-Google-Attributes: gid1089d0,public
From: Joi Ellis <joi@joi.aravox.com>
Subject: Re: Simple problem for float calculation
Date: 2000/10/17
Message-ID: <8shoqj$n2q16@bornews.borland.com>
X-Deja-AN: 682491326
References: <39ec3576_2@dnews>
Organization: Another Netscape Collabra Server User
User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.14-5.0 (i686))
Newsgroups: borland.public.jbuilder.java.api

Reply to Parent Message