<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN">
<html><body>
<p>Dear Eric,</p>
<p>After using gethead and puthead in this way the task CONVL worked perfectly.</p>
<p>Thank you for the explanations.</p>
<p>Regards,</p>
<p> </p>
<p> J. C. Algaba</p>
<p> </p>
<p> </p>
<p>On Mon, 23 Apr 2012 17:02:22 -0600, Eric Greisen wrote:</p>
<blockquote type="cite" style="padding-left:5px; border-left:#1010ff 2px solid; margin-left:5px; width:100%"><!-- html ignored --><!-- head ignored --><!-- meta ignored -->
<pre>Juan-Carlos Algaba-Marcos wrote:</pre>
<blockquote type="cite" style="padding-left:5px; border-left:#1010ff 2px solid; margin-left:5px; width:100%">To whom it may concern, I am lately working on task CONVL with theoretical and observed images. Currently I want to obtain the cross-correlation of a theoretical model and the corresponding synthesized 'observed' one. (OPCODE='IMAC') However, when I try to do so I obtain the following error: localh> CONVL1: CONFIL: UNEQUAL SPACINGS IN TWO MAPS localh> CONVL1: IMAGE= 0.0000 0.0000 CONV. FN= 0.0000 0.0000 localh> CONVL1: Destroyed MA image file: catno= 39 disk= 2 localh> CONVL1: Purports to die of UNNATURAL causes localh> CONVL1: localhost 31DEC11 TST: Cpu= 0.0 Real= 0 As the first image is a model, I have removed the blanks with task REMAG and adjusted all values to match the observed image with PUTHEAD. Here are the headers of both input images:</blockquote>
<pre>IMHEADER does not display the axis increments with enough accuracy to
see what CONVL is complaining about. It checks that the increments are
the same to 5% and they are not. Try keyword='cdelt1'; gethead on one
image and then puthead on the other. Repeat with keyword='cdelt2'.
That should get us past the above error.
It is true that when you convolve with a Gaussian is scales correctly to
produce Jy/beam from Jy/pixel and it should therefore change the header
units. keyword='bunit'; keystr='jy/beam'; puth will fix that. I will
look at correcting the code (in 31DEC12 version).
OHGEO is intended to interpolate an image to the coordinates of the 2nd
image - your two are half-way across the sky from each other and so that
is a meaningless operation (all blanks out results).
Eric Greisen
</pre>
</blockquote>
<div> </div>
</body></html>