[Bug 1992228] Re: Fax/PDF performance regression compared to focal
Julian Andres Klode
1992228 at bugs.launchpad.net
Mon Oct 10 10:29:28 UTC 2022
** Also affects: imagemagick (Ubuntu Kinetic)
Importance: Undecided
Status: New
** Also affects: imagemagick (Ubuntu Jammy)
Importance: Undecided
Status: New
** Changed in: imagemagick (Ubuntu Jammy)
Status: New => Triaged
** Changed in: imagemagick (Ubuntu Kinetic)
Status: New => Triaged
** Changed in: imagemagick (Ubuntu Jammy)
Importance: Undecided => Medium
** Changed in: imagemagick (Ubuntu Kinetic)
Importance: Undecided => High
** Changed in: imagemagick (Ubuntu Jammy)
Importance: Medium => High
--
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to imagemagick in Ubuntu.
https://bugs.launchpad.net/bugs/1992228
Title:
Fax/PDF performance regression compared to focal
Status in imagemagick package in Ubuntu:
Triaged
Status in imagemagick source package in Jammy:
Triaged
Status in imagemagick source package in Kinetic:
Triaged
Bug description:
I'm regularly using `convert` to turn a line-art PNM file to a PDF
with Fax/Group4 compression.
This used to take a few seconds on focal.
I updated to jammy and it takes almost 5 minutes now!
$ time convert -verbose scan.pnm -compress Fax scan.pdf
scan.pnm PBM 5120x7138 5120x7138+0+0 1-bit Bilevel Gray 4.35672MiB 0.610u 0:00.487
scan.pnm=> GROUP4 5120x7138 5120x7138+0+0 1-bit Gray 124981B 284.590u 4:43.594
scan.pnm=> GROUP4 5120x7138=>76x106 76x106+0+0 1-bit Gray 242B 0.150u 0:00.134
scan.pnm=>scan.pdf PBM 5120x7138 5120x7138+0+0 1-bit Bilevel Gray 127042B 285.280u 4:44.267
real 4m44.795s
user 2m5.962s
sys 2m39.970s
Input image here: https://stephan.walter.name/files/scan.pnm
I tried to reproduce the issue with upstream
https://github.com/ImageMagick/ImageMagick6 , but the PDF file
generated was just noise - I'm probably missing some `configure`
options or libraries.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/imagemagick/+bug/1992228/+subscriptions
More information about the foundations-bugs
mailing list