Using thick=2 increases FG plot time by 25x?!?! [message #86592] |
Thu, 21 November 2013 10:35  |
Paul Van Delst[1]
Messages: 1157 Registered: April 2002
|
Senior Member |
|
|
I have the following code to plot some data (~25000 points) using the
PLOT() function:
tic
po = plot(fo, ro, $
xrange=xrange, $
yrange=[-0.005,0.025], $
xticklen = ticklen, $
yticklen = ticklen, $
xtickfont_size = 9, $
ytickfont_size = 9, $
color = 'red', $
thick = 1 ) ; <<<---****
pn = plot(fn, rn, $
color='green', $
symbol = 'diamond', $
/overplot)
pz = plot(po.xrange,[0,0], $
linestyle='dashed', $
/overplot)
pz.order, /send_to_back
toc
When I run this I get the plot and the following output from toc:
% Time elapsed: 1.1067200 seconds.
I then changed *ONLY* the value of thick to "thick = 2", recompiled the
procedure, and ran it. I got the following:
% Time elapsed: 25.625711 seconds.
Am I doing something stupidly wrong here, or is this a known bug?
IDL> print, !version
{ x86_64 linux unix linux 8.2.2 Jan 23 2013 64 64}
cheers,
paulv
p.s. The first time this happened I thought my computer had karked it so
I did the ctrl-C dealy after about 10-15 seconds and got the following:
^C^C^C^C^C^C^C^C
Interrupted at: IDLITVISDATASPACE::DRAW 2389
/usr/local/exelis/idl82/lib/itools/framework/idlitvisdataspa ce__define.pro
IDL> .reset_session
% Interrupted at: IDLITVISDATASPACE::DRAW 2389
/usr/local/exelis/idl82/lib/itools/framework/idlitvisdataspa ce__define.pro
% Internal error: The Interpreter stack is not empty on exit.
|
|
|
Re: Using thick=2 increases FG plot time by 25x?!?! [message #86601 is a reply to message #86592] |
Fri, 22 November 2013 00:17  |
lecacheux.alain
Messages: 325 Registered: January 2008
|
Senior Member |
|
|
Le jeudi 21 novembre 2013 19:35:56 UTC+1, Paul van Delst a écrit :
> I have the following code to plot some data (~25000 points) using the
>
> PLOT() function:
>
>
>
> tic
>
> po = plot(fo, ro, $
>
> xrange=xrange, $
>
> yrange=[-0.005,0.025], $
>
> xticklen = ticklen, $
>
> yticklen = ticklen, $
>
> xtickfont_size = 9, $
>
> ytickfont_size = 9, $
>
> color = 'red', $
>
> thick = 1 ) ; <<<---****
>
> pn = plot(fn, rn, $
>
> color='green', $
>
> symbol = 'diamond', $
>
> /overplot)
>
> pz = plot(po.xrange,[0,0], $
>
> linestyle='dashed', $
>
> /overplot)
>
> pz.order, /send_to_back
>
> toc
>
>
>
> When I run this I get the plot and the following output from toc:
>
>
>
> % Time elapsed: 1.1067200 seconds.
>
>
>
>
>
> I then changed *ONLY* the value of thick to "thick = 2", recompiled the
>
> procedure, and ran it. I got the following:
>
>
>
> % Time elapsed: 25.625711 seconds.
>
>
>
> Am I doing something stupidly wrong here, or is this a known bug?
>
>
>
> IDL> print, !version
>
> { x86_64 linux unix linux 8.2.2 Jan 23 2013 64 64}
>
>
>
>
>
> cheers,
>
>
>
> paulv
>
>
>
>
>
> p.s. The first time this happened I thought my computer had karked it so
>
> I did the ctrl-C dealy after about 10-15 seconds and got the following:
>
>
>
> ^C^C^C^C^C^C^C^C
>
> Interrupted at: IDLITVISDATASPACE::DRAW 2389
>
> /usr/local/exelis/idl82/lib/itools/framework/idlitvisdataspa ce__define.pro
>
>
>
> IDL> .reset_session
>
> % Interrupted at: IDLITVISDATASPACE::DRAW 2389
>
> /usr/local/exelis/idl82/lib/itools/framework/idlitvisdataspa ce__define.pro
>
> % Internal error: The Interpreter stack is not empty on exit.
I cannot reproduce (1.4s on my Workstation in each case)!
{ x86_64 Win32 Windows Microsoft Windows 8.2.3 May 3 2013 64 64}
Maybe were you close to some memory limit in your IDL session, then swapping to disk ?
alx.
|
|
|