bugGNU Octave - Bugs: bug #51141, printing bitmap image to eps is...

 
 

bug #51141: printing bitmap image to eps is slow and produces huge files

Submitted by:  Alois Schlögl <schloegl>
Submitted on:  Tue 30 May 2017 07:49:29 AM UTC  
 
Category: Plotting with OpenGLSeverity: 3 - Normal
Priority: 5 - NormalItem Group: Performance
Status: InvalidAssigned to: None
Originator Name: Alois SchloeglOpen/Closed: Closed
Release: 4.2.1Operating System: GNU/Linux

Add a New Comment(Rich Markup)
   

You are not logged in

Please log in, so followups can be emailed to you.

 

Mon 05 Jun 2017 02:11:24 PM UTC, comment #2:

Closing report.

Pantxo Diribarne <pantxo>
Project Member
Wed 31 May 2017 09:03:21 AM UTC, comment #1:

Hi,

Thanks for your report. "pcolor" produces a surface object not a bitmap image. Now if you look at the eps produced by Matlab you'll see that the surface is replaced by an image.
So unless this behavior is documented somewhere in ML doc I don't think we can call this a bug: Octave produces a vector output from vector objects while ML produces quasi-raster output!

If you are happy with how ML figures appear in your final document (latex??) then there are a couple of workaround in Octave:

  • print to png. This is incredibly slow since Octave will first print to eps and then convert to png, but then the file is 100kB large.
  • use "imagesc" instead of "pcolor"

May I close this report, or do you still think there is a bug?

Pantxo Diribarne <pantxo>
Project Member
Tue 30 May 2017 07:49:29 AM UTC, original submission:

print pixel image to eps is slow and produces huge files, while in Matlab its much faster and produces also smaller files.
The resulting eps files on octave are 100x to 1000x larger without any visible benefit.

-rw-rw-r-- 1 schloegl schloegl 2195862 Mai 30 08:52 fig1.ml2013a.eps
-rw-rw-r-- 1 schloegl schloegl 6521494 Mai 30 08:53 fig1.ml2013a.epsc
-rw-rw-r-- 1 schloegl schloegl 6521606 Mai 30 08:53 fig1.ml2013a.epsc2
-rw-rw-r-- 1 schloegl schloegl 48195 Mai 30 08:52 fig1.ml2016b.eps
-rw-rw-r-- 1 schloegl schloegl 70585 Mai 30 08:52 fig1.ml2016b.epsc
-rw-rw-r-- 1 schloegl schloegl 1247342 Mai 30 08:52 fig1.ml2016b.epsc2
-rwxrw-r-- 1 schloegl schloegl 46405 Mai 30 09:08 fig1.mswin.ml2017a.eps
-rwxrw-r-- 1 schloegl schloegl 69113 Mai 30 09:08 fig1.mswin.ml2017a.epsc
-rwxrw-r-- 1 schloegl schloegl 1225200 Mai 30 09:08 fig1.mswin.ml2017a.epsc2
-rw-rw-r-- 1 schloegl schloegl 100722631 Mai 30 08:55 fig1.octave421.fltk.eps
-rw-rw-r-- 1 schloegl schloegl 100761467 Mai 30 08:59 fig1.octave421.fltk.epsc
-rw-rw-r-- 1 schloegl schloegl 100761467 Mai 30 09:12 fig1.octave421.fltk.epsc2

In order to demonstrate this, I've put up the files at
https://pub.ist.ac.at/~schloegl/20170518428373.matlab.eps.issue/

The script is here:
https://pub.ist.ac.at/~schloegl/20170518428373.matlab.eps.issue/example_eps_issue.m

The data is here:
https://pub.ist.ac.at/~schloegl/20170518428373.matlab.eps.issue/data.mat

I've tested this with fltk.

Alois Schlögl <schloegl>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by pantxo (Posted a comment)
  • -unavailable- added by schloegl (Submitted the item)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only project members can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 3 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Mon 05 Jun 2017 02:11:24 PM UTCpantxoOpen/ClosedOpen=>Closed
    Wed 31 May 2017 09:03:21 AM UTCpantxoCategoryPlotting=>Plotting with OpenGL
      StatusNone=>Invalid

    Back to the top


    Powered by Savane 3.1-cleanup1