Bug 92066 - Calc AVERAGE() Produces incorrect result
Summary: Calc AVERAGE() Produces incorrect result
Status: RESOLVED DUPLICATE of bug 84012
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Calc (show other bugs)
Version:
(earliest affected)
4.3.7.2 release
Hardware: All macOS (All)
: low minor
Assignee: Not Assigned
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-06-14 18:17 UTC by Nathan Koren
Modified: 2015-06-16 02:09 UTC (History)
3 users (show)

See Also:
Crash report or crash signature:


Attachments
Example (8.85 KB, application/vnd.oasis.opendocument.spreadsheet)
2015-06-14 19:29 UTC, Nathan Koren
Details
What I see when I enter the formula (167.88 KB, image/png)
2015-06-14 21:32 UTC, Nathan Koren
Details
Resizing the column a bit... (170.86 KB, image/png)
2015-06-14 21:32 UTC, Nathan Koren
Details
Resizing the column a bit more... finally the correct result! (173.89 KB, image/png)
2015-06-14 21:33 UTC, Nathan Koren
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Nathan Koren 2015-06-14 18:17:01 UTC
=AVERAGE(507, 547, 585) returns 6.3333333...
Comment 1 Nathan Koren 2015-06-14 18:19:06 UTC
Marked this as "critical" because -- simple as it is -- there's no way I can trust my spreadsheets to anything which gets something this fundamental wrong!
Comment 2 Joel Madero 2015-06-14 19:02:40 UTC
Please attach a sample spreadsheet.


Marking as NEEDINFO - once you attach set to UNCONFIRMED.

I just tested on:
Ubuntu 15.04 x64
LibreOffice 5.1 master

works fine - results in 546 1/3

Please also test with 4.4 if possible.
Comment 3 Nathan Koren 2015-06-14 19:29:56 UTC
Created attachment 116530 [details]
Example
Comment 4 Joel Madero 2015-06-14 19:31:37 UTC
Opens fine for me.

Ubuntu 15.04 x64
LibreOffice Version: 4.4.3.2

Please try with 4.4 with a fresh profile as 4.3 is not getting any more updates.
Comment 5 MM 2015-06-14 21:08:48 UTC
Unconfirmed with v4.4.4.1 under windows 7 x64.
Unconfirmed with v4.3.7.2 under mint 17.1 x64.
Comment 6 Nathan Koren 2015-06-14 21:31:28 UTC
Ahhhh... okay, my apologies: I see now that this is not actually a calculation bug, but a display bug. Means the spreadsheet is more trustworthy than I'd feared (whew!) but it definitely is a source of confusion. Basically, the lefthand digits of the result are being cut off in a somewhat arbitrary way. When I first enter the formula, it shows 6.33333... as the result. Narrow the column slightly, it shows .33333... Narrow it a bit more, it shows 546.33333. Am attaching screenshots to demonstrate. Guess the bug description should be changed as well! Also, given the nature of the bug, I would presume this is very OS-specific.
Comment 7 Nathan Koren 2015-06-14 21:32:03 UTC
Created attachment 116534 [details]
What I see when I enter the formula
Comment 8 Nathan Koren 2015-06-14 21:32:29 UTC
Created attachment 116535 [details]
Resizing the column a bit...
Comment 9 Nathan Koren 2015-06-14 21:33:10 UTC
Created attachment 116536 [details]
Resizing the column a bit more... finally the correct result!
Comment 10 Joel Madero 2015-06-14 22:58:04 UTC
Not a critical bug -

Minor - can slow down professional quality work but will not prevent it;
Low - default

Again - would be nice to get you to test 4.4 as 4.3 will not receive any further updates. This might already be fixed in 4.4 for you.
Comment 11 Alex Thurgood 2015-06-15 09:08:23 UTC
This is a DUP, just need to find it...
Comment 12 Alex Thurgood 2015-06-15 09:16:03 UTC

*** This bug has been marked as a duplicate of bug 84012 ***
Comment 13 Alex Thurgood 2015-06-15 09:16:30 UTC
The value is correctly displayed in

Version: 5.1.0.0.alpha1+
Build ID: c0aacd16604548e12d09ca189bb197d0cf5c1afe
Locale : fr-FR (fr.UTF-8)

OSX 10.10.3
Comment 14 Alex Thurgood 2015-06-15 09:17:54 UTC
Also correctly displayed in 

Version: 4.4.3.2
Build ID: 88805f81e9fe61362df02b9941de8e38a9b5fd16
Locale : fr_

OSX 10.10.3
Comment 15 Nathan Koren 2015-06-16 02:09:12 UTC
Tested in 4.4.4.2 on OSX 10.9.5. The problem still occurs after modifying the cell -- but interestingly, if you defocus the window, it corrects itself. So it's a bit less persistent than before. Will test in 5.0 tomorrow.