https://bugs.documentfoundation.org/show_bug.cgi?id=67026

--- Comment #13 from Andrew <[email protected]> ---
I'm going back on my high horse so please bear with me, there is a reason.

The fact that a multiplication or a division can lead to errors in rounding is
a well known problem and requires the use of such as functions as ROUND().

My problem is that, under certain circumstances, addition and subtraction bring
in rounding errors. I understand that numbers are saved in floating point and
that Excel has the same problem but if cell A1 is loaded with a number with
limited decimals (say 15,02) and cell B1 has a value of 10,02 I should expect
to always get A1+B1=25.04 and not 25.03999999999. My main beef is that both
numbers are inputed and not calculated (so I KNOW they have ONLY two decimal
places) and thus output should be ALWAYS on two decimal places.

There is something that would need to be done to how inputed numbers are
memorized, I realize that it will never happen because it would be a major
piece of work (and I no longer have the necessary programming skills). But,
please, no answers such as the one I received "Excel does it too": if somebody
else does something wrong that doesn't justify me repeating the same mistake!

However I would really like to see a more precise piece of documentation which
explains exactly under which conditions does an addition or subtraction of
numbers inputed with a fixed number of decimals give a result that is not
equally exact.

-- 
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Libreoffice-bugs mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

Reply via email to