After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 127287 - Bad guess for remaining time
Bad guess for remaining time
Status: RESOLVED OBSOLETE
Product: epiphany
Classification: Core
Component: Downloads
2.14.x
Other Linux
: Normal minor
: ---
Assigned To: Epiphany Maintainers
Marco Pesenti Gritti
Depends on: 618443
Blocks:
 
 
Reported: 2003-11-18 12:51 UTC by Edd Dumbill
Modified: 2011-03-07 20:45 UTC
See Also:
GNOME target: ---
GNOME version: 2.13/2.14



Description Edd Dumbill 2003-11-18 12:51:20 UTC
Fwd'd from http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=218480

From: "Marcelo E. Magallon" <mmagallo@debian.org>

Hi,

 I paused a download because it was hogging the downstream connection.
 After a while, I resumed it again... Epiphany seems to remember only
 the time when it started the download, which means it guesses the
 remaining time as:

            t = (S0 - S)/r

 where r is download rate, and S0 the total size, S whatever it has
 downloaded thus far.  And r seems to be:

            r = S/T

 where T is the enlapsed time.  This means that when I pause the
 download, this rate doesn't have anything to do with the current
 download rate but with some sick average with no physical significance
 at all.  Using the current download speed is also bad, because that can
 change wildly.  Using the moving average over the last N time steps
 produces the desired result.

            r = S'/T'

 where S' is the ammount of information downloaded in the last N
 timesteps and T' is the corresponding time.

 Now, if instead of reinventing the weel, Epiphany were to use something
 like GTM we wouldn't be having this discussion.  Sadly the development
 of GTM seems to have stalled, it hasn't been ported to GNOME 2 which
 means it's not cool anymore... there's something newer, downman, but it
 doesn't seem to be packaged for Debian yet.
Comment 1 John Spray 2004-09-20 19:59:32 UTC
One way to deal with this case would be to play a trick on the download:
remember the time at which the pause button was clicked, and then when the
resume button is clicked advance the "time at which the download was started"
variable by how long the download was paused for.

I can't see how to do this neatly, since I don't quite understand the code (the
implementation is somewhere in mozilla, or I'm looking in the wrong place).
Comment 2 Christian Persch 2004-10-13 10:51:51 UTC
Mass reassigning of Epiphany bugs to epiphany-maint@b.g.o
Comment 3 Reinout van Schouwen 2006-01-14 00:33:14 UTC
Edd, is this problem still reproducable in the latest Epiphany version?
Comment 4 Christian Persch 2006-02-07 20:03:11 UTC
ff uses an average of the last 10 data points, see
https://bugzilla.mozilla.org/show_bug.cgi?id=245725 .
Comment 5 Josselin Mouette 2006-05-13 19:01:50 UTC
Reinout: yes, this is still here in 2.14.
Comment 6 Reinout van Schouwen 2006-05-14 10:07:41 UTC
Updating version fields.
Comment 7 Diego Escalante Urrelo (not reading bugmail) 2011-03-07 20:45:33 UTC
Can't see it anymore in the new downloads UI.