| Postscript-problem... :(( [message #10288] |
Thu, 13 November 1997 00:00  |
rew032
Messages: 14
Registered: July 1996
|
Junior Member |
|
|
Hello!
We have a problem with the postscript output......
First we produce a plot on the screen, e.g. a xy-plot.
Than, with a button, we do this:
'GTest' : $
BEGIN
TVLCT, r, g, b, /GET
Pict = TVRD()
WRITE_GIF, 'plottest.gif', Pict, r, g, b
print, 'Bild wurde als plottest.gif abgespeichert.'
SET_PLOT, 'ps'
DEVICE, /COLOR
TV, Pict
DEVICE, /CLOSE
SET_PLOT, 'x'
print, 'PS auch'
the gif-file is 'very' nice, but no good resolution for printing.
therefor we store an additional ps-file.
it has a nice resolution i think, but it shows not the xy-plot,
only a orange rectangular instead..... :(((((
can anyone help?
regards, astrid kuhr
|
|
|
|
| Re: Postscript-problem... :(( [message #10354 is a reply to message #10288] |
Tue, 18 November 1997 00:00  |
Helge.Rebhan
Messages: 12
Registered: October 1997
|
Junior Member |
|
|
Hi,
In article <MPG.eda68b7328deb7f9896aa@news.frii.com>, davidf@dfanning.com
(David Fanning) wrote:
>
> Exactly. This is certainly what was happening to Astrid Kuhr. That
> is why I advised him to get the color table vectors *before*
> he went into the PostScript device:
>
> TVLCT, r, g, b, /Get
>
> And to restore them *exactly* the way he found them once
> he got there:
>
> Set_Plot, "PS"
> TVLCT, r, g, b
>
just one question on this: Does these object orienented graphics stuff in
IDL 5.0
change anything to this old printing device problem ?
Servus, Helge
--
Adjust e-mail address for reply
|
|
|
|
| Re: Postscript-problem... :(( [message #10357 is a reply to message #10288] |
Mon, 17 November 1997 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Joel D. Offenberg (offenbrg@fondue.gsfc.nasa.gov)
offers some good advice when he writes:
> davidf@dfanning.com (David Fanning) writes:
>
>>> - something goes wrong with the colormap.
>>> on the screen I have colors from blue to red, on the ps only from
>>> blue to yellow...??
>>> whats going wrong..?
>
>> Uh, probably you didn't do *exactly* what I suggested in my
>> earlier message and reload those color vectors after you got
>> into the PostScript device. :-)
>
> Actually, I've seen this before---when you draw things in a TV window and
> use the color tables, the output is truncated to the number of colors
> available (which is typically less than 256, since some will be taken up
> by the system, unless you have a 24-bit graphics card). However, that isn't
> the case when you are using PostScript, so things may end up non-linear.
Exactly. This is certainly what was happening to Astrid Kuhr. That
is why I advised him to get the color table vectors *before*
he went into the PostScript device:
TVLCT, r, g, b, /Get
And to restore them *exactly* the way he found them once
he got there:
Set_Plot, "PS"
TVLCT, r, g, b
Combined with his TVRD(), this would result in PostScript
colors EXACTLY like he saw on the display.
> The test is to see if "Print, !d.N_colors" before you start your tvrd gives
> you 256. If it doesn't, you have several options to get it right. The
> easiest way is to pad out the R,G,B vectors to have 256 elements each,
> then "tv, IMAGE < n_colors"
> where n_colors = !d.N_Colors from before you started with the TVRD.
I'm not sure this is the best way if you want to view
your data correctly. I think the best way is usually to scale
your data to the number of colors you have on the display
and make sure you use *exactly* the same color table vectors
(i.e., with same number of elements) both on the display and
in your PostScript file.
Cheers,
David
P.S. In any case, Astrid told me he got better results when
he followed my directions more closely. :-)
-----------------------------------------------------------
David Fanning, Ph.D.
Fanning Software Consulting
E-Mail: davidf@dfanning.com
Phone: 970-221-0438
Coyote's Guide to IDL Programming: http://www.dfanning.com/
|
|
|
|
| Re: Postscript-problem... :(( [message #10359 is a reply to message #10288] |
Mon, 17 November 1997 00:00 |
offenbrg
Messages: 31
Registered: August 1993
|
Member |
|
|
davidf@dfanning.com (David Fanning) writes:
> A.Kuhr (rew032@isr141.isr.kfa-juelich.de) writes:
>> Thank you very much, now not only an orange rectangle... :))))
>>
>> But 2 other problems occured.... :(((
>>
>> - the resolution of the printed postscript is very :((( to, like gif.....
>> how get a 'nicer' picture...?
> Yes, it is a *screen dump* you are doing. Resolution is
> VERY low compared to true PostScript output. To get a nicer
> PostScript picture re-issue the exact same commands that got
> you that nice picture on your display. It would help if those
> commands were in a separate procedure. :-)
>
>> - something goes wrong with the colormap.
>> on the screen I have colors from blue to red, on the ps only from
>> blue to yellow...??
>> whats going wrong..?
> Uh, probably you didn't do *exactly* what I suggested in my
> earlier message and reload those color vectors after you got
> into the PostScript device. :-)
Actually, I've seen this before---when you draw things in a TV window and
use the color tables, the output is truncated to the number of colors
available (which is typically less than 256, since some will be taken up
by the system, unless you have a 24-bit graphics card). However, that isn't
the case when you are using PostScript, so things may end up non-linear.
The test is to see if "Print, !d.N_colors" before you start your tvrd gives
you 256. If it doesn't, you have several options to get it right. The
easiest way is to pad out the R,G,B vectors to have 256 elements each,
then "tv, IMAGE < n_colors"
where n_colors = !d.N_Colors from before you started with the TVRD.
Joel
--
"...And I am unanimous in this" - Mrs. Slocumbe
------------------------------------------------------------ ---------------
| Joel D Offenberg | Joel.D.Offenbrg.1@gsfc.nasa.gov |
| Hughes STX, NASA/GSFC/LASP | UIT, COBE, NGST programmer & sysadmin |
|
|
|
|
| Re: Postscript-problem... :(( [message #10402 is a reply to message #10288] |
Tue, 25 November 1997 00:00 |
Michael Werger
Messages: 34
Registered: May 1997
|
Member |
|
|
A.Kuhr wrote:
> Hello!
>
> What I heard, not still now tried by myself, if you use
> the object orientated things, it is easier.
> but next problem: no vector postscript is generated.
> instead of this pixel....!!
> and than you get some 20mb or more postscript files....
> good luck for you spooling filesystem and yor printer....
>
> regards, astrid
As far as I can remember: this was also one of the complaints on the oo
graphics at the
ADASS conference this September. David (Stern) told us (the users)
that this "feature" will
probably changed in the future - we'll see
--
Michael Werger ESA ESTEC & Praesepe B.V.
Astrophysics Division mwerger@estec.esa.nl
Postbus 299 http://astro.estec.esa.nl
2200 AG Noordwijk +31 71 565 3783 (Voice)
The Netherlands +31 71 565 4690 (FAX)
|
|
|
|
| Re: Postscript-problem... :(( [message #10416 is a reply to message #10288] |
Mon, 24 November 1997 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Astrid Kuhr (rew032@isr141.isr.kfa-juelich.de) writes:
> What I heard, not still now tried by myself, if you use
> the object orientated things, it is easier.
Not exactly true. If you use object-oriented graphics
routines you can use the PRINTER device directly. But
you can also take advantage of the PRINTER device with
the normal, direct graphics routines.
> but next problem: no vector postscript is generated.
> instead of this pixel....!!
No, the PRINTER device will draw vectors. It is just
that object graphics are in a true 3D environment, so
that even something simple like a line in a plane has
a 3D component that must be passed around with it. This,
necessarily, slows things down a bit and makes the file
larger.
> and than you get some 20mb or more postscript files....
> good luck for you spooling filesystem and yor printer....
I don't think direct graphics commands sent to the PRINTER
device will generate anything like this. I don't usually
find myself complaining about the speed of PRINTER output,
only about the implementation. I wish the PRINTER device
was more like the PS device for PostScript output. In
particular, I have found it impossible to display any
graphics output that has an image on it in addition to
other graphics commands. This is due to a lack of control
with respect to how graphics are positioned and sized in
the PRINTER device.
Cheers,
David
-----------------------------------------------------------
David Fanning, Ph.D.
Fanning Software Consulting
E-Mail: davidf@dfanning.com
Phone: 970-221-0438
Coyote's Guide to IDL Programming: http://www.dfanning.com/
|
|
|
|
| Re: Postscript-problem... :(( [message #10418 is a reply to message #10288] |
Mon, 24 November 1997 00:00 |
rew032
Messages: 14
Registered: July 1996
|
Junior Member |
|
|
Hello!
What I heard, not still now tried by myself, if you use
the object orientated things, it is easier.
but next problem: no vector postscript is generated.
instead of this pixel....!!
and than you get some 20mb or more postscript files....
good luck for you spooling filesystem and yor printer....
regards, astrid
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
