[Thread Prev][Thread Next][Index]

Re: [ferret_users] plot/vs and axes



Hi Ryo-
Yes, I appreciate the work it takes to get the right solution to some of these detailed problems. We have run into the issue of the size of the plot object ourselves, many times.

But also, stay tuned. We are working now on implementing Ferret with a new graphics library, not based on xgks, but a newer modern graphics library. The Ferret syntax and the graphical output will look the same, or nearly so, but we will have the option of saving plots in a number of graphics formats including pdf and postscript and png. The long-standing issue of the white lines (which makes many appearances in these archives, as in http://www.pmel.noaa.gov/maillists/tmap/ferret_users/fu_2010/msg00630.html) will go away. And it will let us use other fonts, though perhaps not in the first release of this. Finally, because it does not depend on X, it should be possible to have a new Ferret under Windows.

Ansley

On 11/22/2011 3:29 PM, Ryo Furue wrote:
Hi Ansley,

Thank you for your response.

| but doing a shade plot first isn't really much of a
| burden, is it?

No, not any longer. :-)  I mean, I had to think a lot
and do some research before arriving at a satisfactory
solution.   For example, I didn't know the existence
of the NODATA_LAB feature.

The thing is, I want to minimize the number of objects
generated into the final PostScript file.  Doing the
same SHADE twice would double the number of objects.

One drawback of that is the size of the resultant
PostScript file.

I usually create my presentations in PDF, into which
I import PostScript figures.  So, I often find my
presentation so "heavy" that the PDF viewer renders
it too slowly, which is painful.  This happens when
I include plots from high-resolution data (like that
of an eddy-resolving model).

And then the PostScript editor I use is so slow
that it's painful to edit a big file.

So, simply doing the same SHADE twice isn't a good
solution to me.  I then came up with this solution:

   let nothing = temp + 0/0 !! all missing
   cancel mode NODATA_LAB
   SHADE/ . . .  nothing  !! just to plot the axes.

In place of this, you can also plot a variable with
an out-of-range LEVEL.

| First, a small thing, in your script, you have
|
|   ! cancel NODATA_LAB  !! doesn't work on my Ferret version.
|
| That should be cancel MODE NODATA_LAB.

Ah, that was my mistake.  Yes, as you suspected,
I just forgot the word "MODE" and thought my version didn't
support it.  Thanks for correcting me.

Regards,
Ryo


[Thread Prev][Thread Next][Index]
Contact Us
Dept of Commerce / NOAA / OAR / PMEL / Ferret

Privacy Policy | Disclaimer | Accessibility Statement