comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » Re: mixing plots and images in IDL
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Switch to threaded view of this topic Create a new topic Submit Reply
Re: mixing plots and images in IDL [message #5019] Tue, 12 September 1995 00:00
m.c.harrison is currently offline  m.c.harrison
Messages: 5
Registered: July 1995
Junior Member
lucio@ifctr.mi.cnr.it (Lucio Chiappetti) wrote:
>
> I had been told that the correct way to plot an image in colour or
> grayscale in an axis frame (with or without overplotted a contour) was :
>
> ; set up ps device
> ;

>
> device,/helvetica,font_size=12
> plot axes
> tv main image
> replot axes
> tv colour bar with fudge
> device,/helvetica,font_size=5
> annotate tics
>
> IS WRONG
>
> device,/helvetica,font_size=5 SIC !
> plot axes
> tv main image
> replot axes
> tv colour bar with fudge
> device,/helvetica,font_size=5
> annotate tics
>
> IS CORRECT (BUT NOT WHAT I WANT !)
>
> device,/helvetica,font_size=5
> plot axes
> tv main image
> tv colour bar with fudge
> device,/helvetica,font_size=5
> annotate tics
> device,/helvetica,font_size=12
> replot axes
>
> GIVES THE CORRECT COLOUR BAR BUT THE AXES ARE REPLOTTED WRONG
>
> Now I'd try to play around always with /NORM coordinates, but I'd appreciate
> if somebody could explain the interference between TV, axis setting and font
> setting ....
I had this sort of problem attempting to make my contour mapping routine
more generic than just drawing to a window. The problem lay with
tv when creating a colour bar. Data coordinates appeared quite
different, stretchin over my map and causing mayhem and so I created
a box using normal coords, found the extents in data space using
conv_coords and filled it that way. THis has problem that anything
large enough to be useful can cause the machine to run out of memory,
since examining !d.whatsit with 'ps' device set shows there to be
hundreds of data points per cm. Eventually I used poly_fill
which avoids this problem altogether and might be suitable since
youre only byte scaling ..
Hope this is a help
Mike
m.c.harrison@rl.ac.uk
Re: mixing plots and images in IDL [message #5020 is a reply to message #5019] Tue, 12 September 1995 00:00 Go to previous message
m.c.harrison is currently offline  m.c.harrison
Messages: 5
Registered: July 1995
Junior Member
lucio@ifctr.mi.cnr.it (Lucio Chiappetti) wrote:
>
> I had been told that the correct way to plot an image in colour or
> grayscale in an axis frame (with or without overplotted a contour) was :
>
> ; set up ps device
> ;

>
> device,/helvetica,font_size=12
> plot axes
> tv main image
> replot axes
> tv colour bar with fudge
> device,/helvetica,font_size=5
> annotate tics
>
> IS WRONG
>
> device,/helvetica,font_size=5 SIC !
> plot axes
> tv main image
> replot axes
> tv colour bar with fudge
> device,/helvetica,font_size=5
> annotate tics
>
> IS CORRECT (BUT NOT WHAT I WANT !)
>
> device,/helvetica,font_size=5
> plot axes
> tv main image
> tv colour bar with fudge
> device,/helvetica,font_size=5
> annotate tics
> device,/helvetica,font_size=12
> replot axes
>
> GIVES THE CORRECT COLOUR BAR BUT THE AXES ARE REPLOTTED WRONG
>
> Now I'd try to play around always with /NORM coordinates, but I'd appreciate
> if somebody could explain the interference between TV, axis setting and font
> setting ....
I had this sort of problem attempting to make my contour mapping routine
more generic than just drawing to a window. The problem lay with
tv when creating a colour bar. Data coordinates appeared quite
different, stretchin over my map and causing mayhem and so I created
a box using normal coords, found the extents in data space using
conv_coords and filled it that way. THis has problem that anything
large enough to be useful can cause the machine to run out of memory,
since examining !d.whatsit with 'ps' device set shows there to be
hundreds of data points per cm. Eventually I used poly_fill
which avoids this problem altogether and might be suitable since
youre only byte scaling ..
Hope this is a help
Mike
m.c.harrison@rl.ac.uk
  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: Plot titles
Next Topic: Wanted : Surface Lifted Index Routine

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Fri Oct 10 09:58:24 PDT 2025

Total time taken to generate the page: 0.64398 seconds