octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Octave-gui graphics problems


From: Vic Norton
Subject: Re: Octave-gui graphics problems
Date: Mon, 13 Jun 2016 09:09:40 -0400

> On Jun 12, 2016, at 11:27 PM, Dmitri A. Sergatskov <address@hidden> wrote:
> 
> 3) Printing of the plots with many points takes long time. This is an issue 
> with gl2ps.
> We could discuss what to do about that. One mitigation strategy would be to 
> bypass
> gl2ps when one saves to bitmap files (like png).

The times I reported (below) have nothing to do with saving the results to any 
kind of file. Saving a computed graphic to a PDF to takes almost no time with 
gnuplot-qt. The big difference is in the size of the saved file. (Saving the 
straight-qt graphic is a waste for me. The saved graphic is REALLY BAD on my 
iMac system.)

I just recomputed my example
  qtbug_gnuplot-qt_15000.120.png (69.9 KB)
and saved the result to PDF. This produced
  qtbug_gnuplot-qt_15000.120.pdf (563.2 KB)
The PDF looks MUCH better than the PNG. The blue dots in the PDF are smaller 
and perfectly round. The text and lines are sharp and crisp. The real cost is 
in the size of the file, not in the time to produce it.

I won’t post the above PDF. It is way too big for this list and for 
publication. Here is what I’ll be using for publication. It’s the PNG 
background with LaTeX/TikZ lines and annotation. It only weighs 83.8 KB.

Attachment: NM_XX.pdf
Description: Adobe PDF document



> On Jun 12, 2016, at 5:29 PM, Vic Norton <address@hidden> wrote:
> 
>  gnuplot-qt
>    nR = 10000, mean = 0.32 sec, stdev = 0.02
>    nR = 15000, mean = 0.39 sec, stdev = 0.02
>    nR = 20000, mean = 0.46 sec, stdev = 0.02
>    nR = 25000, mean = 0.52 sec, stdev = 0.01
>    nR = 30000, mean = 0.59 sec, stdev = 0.01
>  qt
>    nR = 10000, mean = 0.36 sec, stdev = 0.03
>    nR = 15000, mean = 0.40 sec, stdev = 0.01
>    nR = 20000, mean = 0.46 sec, stdev = 0.02
>    nR = 25000, mean = 0.56 sec, stdev = 0.04
>    nR = 30000, mean - 0.62 sec, stdev = 0.04

Regards,

Vic

reply via email to

[Prev in Thread] Current Thread [Next in Thread]