see bellow.

2009/2/5 Marc Schwartz <[email protected]>

> Ely Schoenfeld <[email protected]> writes:
>
> > Hello all.
> >
> > I discovered something odd with OOo 3.0.1 Calc.
> >
> > Hope someone could verify this for me.
> >
> > a) Put in two different cells the numbers:
> >
> > 18282.34   (For example A1)
> > -18282.73  (For example A2)
> >
> > b) Add them in another cell. (For example A3)
> > Doesn't matter if you use the "SUM" formula or just use a simple "+"
> sign.
> >
> > If I add more decimals to the cell format, I get something like
> -0.390000,
> > that's ok.
> >
> > But If I copy the last cell and then paste special with only text,
> numbers and
> > dates checked, I obtain:
> > -0.389999999999418
> >
> > You can test that the problem IS affecting further calculations like
> this.
> >
> > a) Add 0.005 to the result (A3 in my example)
> >
> > You will obtain -0.38 if you are showing 2 decimals.
> >
> > b) Write exactly -0.39 in a new cell, and then add the same 0.005 to it.
> >
> > You will obtain -0.39 if you are showing 2 decimals.
> >
> > In both cases, if you add more decimals to the cells, you will obtain the
> same:
> > -0.385
> > But if you copy and paste special with only text, numbers and dates
> checked,
> > you would obtain in the first case -0.384999999999418 and -0.385 in the
> second
> > one.
> >
> > I'm attaching an odt file here, hope it passes through.
> >
> > In case it's relevant, this happened on a:
> >
> > Windows XP Version 2002, sp3
> > 32Bit Centrino Duo processor
> >
> > Should I open an issue about this? (Couldn't find one)
> >
> > Thank you all.
> >
> > Ely.
>
> In addition to all of the other comments, you might want to review this
> article:
>
> What Every Computer Scientist Should Know About Floating-Point
> Arithmetic
>
> http://docs.sun.com/source/806-3568/ncg_goldberg.html
>
>
> You might also want to review this MSKB article:
>
> http://support.microsoft.com/default.aspx?scid=kb;EN-US;78113
>
> as OO.org has implemented a similar process in Calc.
>
>
> HTH,
>
> Marc Schwartz
>

Thank you very much for those links.

Does anybody knows if this Floating Point error gets moved to a much bigger
numbers if I install OOo Calc in a 64bit computer with a 64bit compatible
OS? Or it depends on the format it uses internally?

Ely


>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>

Reply via email to