[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Octave-bug-tracker] [bug #61647] Plot performance tic/toc misleading re
From: |
anonymous |
Subject: |
[Octave-bug-tracker] [bug #61647] Plot performance tic/toc misleading result |
Date: |
Sat, 11 Dec 2021 00:10:01 -0500 (EST) |
User-agent: |
Mozilla/5.0 (Windows NT 6.1; rv:94.0) Gecko/20100101 Firefox/94.0 |
Follow-up Comment #2, bug #61647 (project octave):
Not only resizing the figure leads to re-rendering but moving the figure or
minimizing and maximizing(without size change) causes re-rendering on qt
toolkit (maybe subject of another issue?) .
I also used the following command:
tic;plot(a, b, '.');drawnow;toc
Now fltk waits until the rendering is finished and, to my opinion, shows the
correct timing but gnuplot and qt don't wait for rendering. Actually the
command window is frozen while rendering but the timing doesn't include
rendering process.
Using clf:
tic;plot(a, b, '.');clf;toc
Nothing is drawn (according to CPU usage). In this case the last command is
executed.
Using drawnow+clf:
tic;plot(a, b, '.');drawnow;clf;toc
Voila! It prints correct timings for both qt and fltk.
All of the above timings as noted in comment #1 my be the expected behavior
but I have seen in several bug reports that the performance of plot is
measured by tic/toc.
Using [tic + plot + drawnow + clf + toc] wonderfully shows that on my computer
fltk is more performant than qt (at least in plotting points)!
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?61647>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/