[Libreoffice-bugs] [Bug 126882] When a column's width is increased, the subraction-result in it is off by a little.

[bugzilla-daemon at bugs.documentfoundation.org]

https://bugs.documentfoundation.org/show_bug.cgi?id=126882

multipartite at gmail.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
     Ever confirmed|0                           |1
         Resolution|NOTABUG                     |---

--- Comment #3 from multipartite at gmail.com ---
(In reply to Oliver Brinzing from comment #2)
> this is not a bug, this is how floating point arithmetic works.
> 
> please have a look at:
> 
> Bug 116853 - Simple Subtraction Error Precision
> Bug 126552 - calc subtraction imprecision

Thank you for your feedback!  I'm sorry if I'm setting the wrong Status, or if
I shouldn't have changed the Status as I write the below question--from (
https://wiki.documentfoundation.org/QA/Bugzilla/Fields/Status/REOPENED ) I
thought I should set this to 'NEEDINFO', but I could not see a NEEDINFO option
in the list.

I indeed had not realised that the linked bug reports already existed. 
However, if I understand correctly, that problem described should only be a
function of Binary Floating-Point, where a base-10 number is inexactly
represented as a base-2 number.  As base 10 is being used for spreadsheet
calculations, shouldn't Decimal Floating-Point be used to represent the digits
exactly (within the limits of significance, and without making errors pop up in
unrelated orders of magnitude)?

https://en.wikipedia.org/wiki/Decimal_floating_point

(It is plausible that I am very confused.)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/libreoffice-bugs/attachments/20190814/ba1f3378/attachment.html>