[alma-config]

[Min Yun]

Hi John,

We can basically go about this in two different ways.  One way is
how I propose, which is to try to image the same sources with 0.01"
scale structures using three different resolution configurations.
Another is to change the header so that all of the images have
the same resolution elements acorss, 1024 x 1024.  My reasoning
for taking the first path rather than the second is that the
number of resolution elements within a primary beam drops off
quickly as you go to lower resolution arrays and end up having
most of the model images being irrelevant (except for mosaic 
cases) in the latter approach.
If we follow the second path, we also have to triple the number
of model images to keep around (one for each resolution array).  
I don't any strong preference for one over the other, but 
keeping the same input images for all resolutions seem to make 
the bookkeeping easier.  Any other thoughts on this point?




					-- Min
					
>
------------- Begin Forwarded Message -------------

Date: Tue, 25 Jul 2000 14:34:15 +0200 (CEST)
From: John Conway <jconway at ebur.oso.chalmers.se>
To: Min Yun <myun at zia.aoc.NRAO.EDU>
cc: myun at zia.aoc.NRAO.EDU, heddle at totalise.co.uk, asw at roe.ac.uk
Subject: Re: [alma-config] 
MIME-Version: 1.0


Hi,

 Just droppped in to read emails. Read this on the simulation
web page.


'On July 20, 2000, all model images are modified to have 1024 x 1024
pixels
in order to run on the AIPS simulation package S. Heddle is
putting together. All images have 0.01" pixels. For lower resolution
simulations, comparisons should be made with convolved images
appropriate for the simulation observations. '

Surely what you mean is that the pixel size is to be changed to
be relevant to each resolution of array. i.e. you don't convolve 
them down, just change the effective size using PUTHEAD in 
AIPS, to make images relevant for C and E array. This way 
in each array one can  have 4 or 5 pixels per beam and 50-200
beams across the image as we agreed in Tuscon.

At each resolution of course one much produce a perfect convolved
down image (convolved with the restoring beam) to be compared to
the AIPS output.

  Cheers
   John,