<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Here is sample output from displaying slp after <br>
'set gxout stat'<br>
ga-> set gxout stat<br>
ga-> d slp<br>
Data Type = grid<br>
Dimensions = 0 1<br>
I Dimension = 1 to 289 Linear 0 1.25<br>
J Dimension = 1 to 181 Linear -90 1<br>
Sizes = 289 181 52309<br>
Undef value = 9.999e+20<br>
Undef count = 0 Valid count = 52309<br>
Min, Max = 93261.6 103502<br>
Cmin, cmax, cint = 94000 103000 1000<br>
Stats[sum,sumsqr,root(sumsqr),n]: 5.27395e+09 5.31859e+14
2.30621e+07 52309<br>
Stats[(sum,sumsqr,root(sumsqr))/n]: 100823 1.01676e+10 100835<br>
Stats[(sum,sumsqr,root(sumsqr))/(n-1)]: 100825 1.01678e+10 100836<br>
Stats[(sigma,var)(n)]: 1542.32 2.37875e+06<br>
Stats[(sigma,var)(n-1)]: 1542.33 2.3788e+06<br>
<br>
This tiny bit of grads script, if run immediately after you<br>
display when 'set gxout stat' is set, should help you capture<br>
the line with the min and max, and then assign values<br>
to min1 and m ax1. Then you write a script to loop through<br>
and compare min and max throughout the time series and<br>
use that to set the contour levels...<br>
<br>
rec = sublin(result,9)<br>
min1 = subwrd(rec,5)<br>
max1 = subwrd(rec,6)<br>
<br>
<br>
Dizzle Man wrote:
<blockquote
cite="mid1fbbfe0f0808061519o754813aeob486895298c21dbb@mail.gmail.com"
type="cite">
<pre wrap="">Excellent!, I have tried option two and that works fine. There is one slight
issue though. Is it possible to remove the numbers that come up on the
display when I use contour plotting. They kind of get in the way when I am
observing my data. Also in option 1, what do the min and max represent in
set gxout stat?
On Tue, Aug 5, 2008 at 2:38 PM, Austin Conaty <a class="moz-txt-link-rfc2396E" href="mailto:Austin.L.Conaty@nasa.gov"><Austin.L.Conaty@nasa.gov></a>wrote:
</pre>
<blockquote type="cite">
<pre wrap="">One possible solution is to
1) set gxout stat
2) loop through the data and capture min and max
3) using the min/max info, use the same set ccols and
set clevs commands before each display
4) set gxout shaded and plot.
Or, if you're happy with the color and shading used in the first time
period
another possible solution is to
1) set t 1
2) set gxout shaded
3) d relvort
4) q shades
5) set ccols and set clevs according to info returned
by q shades.
Eric Altshuler wrote:
</pre>
<blockquote type="cite">
<pre wrap="">That is odd because in my experience it is in animations that contouring
and color schemes can be ill-suited to the data being plotted. That's
because whatever contouring scheme is used (or chosen by grads) for the
first frame, is also applied to all the remaining frames so as to ensure the
same contouring scheme for the whole animation. This can be a problem,
though, if the data field being animated changes a lot during the animation
(e.g. a rapidly intensifying hurricane).
When your are plotting each time step individually, grads chooses the
contouring/color scheme separately for each step unless you set it yourself.
In this situation, if the character of the data field changes significantly
in time, grads will choose new contouring schemes to suit the data at each
time step. The disadvantage here is that the contouring scheme is not
consistent across all frames. It's a tradeoff.
How are you doing your animations? Are you setting T to vary and plotting
the variable, or have you set up a script loop to display one time step,
sleep for some interval, advance T and then display the next time step? The
behavior of animations using these two methods can be quite different.
----- Original Message -----
From: "Dizzle Man" <a class="moz-txt-link-rfc2396E" href="mailto:selfscience06@GMAIL.COM"><selfscience06@GMAIL.COM></a>
To: <a class="moz-txt-link-abbreviated" href="mailto:GRADSUSR@LIST.CINECA.IT">GRADSUSR@LIST.CINECA.IT</a>
Sent: Saturday, August 2, 2008 1:28:03 PM GMT -05:00 US/Canada Eastern
Subject: Relative Vorticity Contouring
Hello Everyone,
I am doing a plots for relative voricity [hcurl(u,v)]. The issue that I
am
having is when I run my animation I get a beautiful sequence of images, so
when I begin the plot each time step individually I do not receive that
same
contouring intervals that is displayed in the animation. Its like the
individual plotting loses coloring, but that animation gives me what I am
looking for. Does this make sense? I was wondering how can I maintain the
color sequence that the animation gives for individual time step plotting?
</pre>
</blockquote>
</blockquote>
<pre wrap=""><!---->
</pre>
</blockquote>
<br>
</body>
</html>