Re: Maximized rectangular graphic/plot

From: Dave Allured - NOAA Affiliate <dave.allured_at_nyahnyahspammersnyahnyah>
Date: Sun Mar 03 2013 - 12:14:15 MST

Paulo,

Use a post processing tool to remove the white borders, such as this
ImageMagick command. This is a Unix/Linux command command line, not
an NCL command:

   convert -trim +repage ncl_output.png final.png

The final aspect ratio is determined indirectly by the visible details
in your NCL graphic, not directly through any single resource. The
specifications of the X and Y axes are a starting point. However, the
final PNG aspect ratio has to accommodate the position of titles, axis
labels, color bar, and other annotations, as well as the main figure.
I have found that a trial and error approach gives good results with
very few iterations, once you get used to the post processing command.

To increase the pixel resolution in the final PNG, increase
wks@wkWidth and wks@wkHeight together. Don't try to do this with a
convert command option, because that will simply make a blurry picture
with more pixels! Be careful, there is a strong order-2+ tradeoff
with memory consumption and program speed when @wkWidth and @wkHeight
are increased.

--Dave
Reply to list only!

On Sun, Mar 3, 2013 at 8:09 AM, Paulo Santiago <paulohsm@gmail.com> wrote:
> Dear NCL users,
>
> I'm wondering how to get a graphic in PNG format with 3:4 aspect ratio (or
> any other than 1:1) in a maximized fashion (that is, with zero or near to
> zero space between the plot limits and the canvas edges).
>
> I changed the canvas aspect ratio by setting @wkWidth and @wkHeight of my
> wks. When I use res@gsnMaximize=True, I got a rectangular canvas but square
> graphic.
>
> Using res@vpHeightF and res@vpWidthF I control the aspect ratio of my
> graphic, but I could not get such maximized graphic (even if I use
> res@gsnMaximize=True).
>
> I also changed res@gsnBoxMargin to zero. There are always large blank strips
> at upper and lower regions.
>
> I would like to know if there are more resources I could be missing and what
> resources would these be.
>
> Thanks in advance!
>
> Cheers,
>
> --
> Paulo Santiago
> Ph.D. Student in Meteorology
> Brazilian National Institute for Space Research
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Sun Mar 3 12:14:26 2013

This archive was generated by hypermail 2.1.8 : Thu Mar 07 2013 - 08:55:58 MST