[daip] Problems averaging channels.

[Eric Greisen]

Craig Walker wrote:
> I would like to get a POSSM output data file without as many channels as 
> I'm getting now by averaging channels.  I have tried two methods, both 
> with problems.
> 
> First AVSPC:
> default avspc
>    ind 3; getn 71
>    outn=inn; outcl='avspc'; outs=ins; outd=ind
>    docalib=1; gainuse=maxtab('CL'); flagver=maxtab('FG')
>    doacor=1; channel=3; avoption='subs'
>    smooth=1,0
> 
> That takes in a 2150 vis file and turns it into a 24 vis file with only 
> the autocorrelations from one station.
> 
> Setting DOACOR=-1 left an empty file.
> 
> SPECR:
> 
> default specr
>    ind 3; getn 71
>    outn=inn; outcl='specr'; outs=ins; outd=ind
>    npoints=1024
> 
> Seems to give a good file, but with the sort order blank which causes 
> POSSM a fit.  I can either SORT or change the header, but that's a bit 
> awkward.  I don't think it changed the vis records so I'm not sure why 
> the sort order changed.
> 
> This is all on noatak at the aoc, disk 3.

I found that OOP objects set the sort order blank when they create the 
file and the order has to be forced into the header after that (not 
before as the task did already).  I also fixed a bug in handling old 
format FQ tables.  More for tonight's MNJ!

I wonder if the avspc may be due to a calibration issue - did you try 
those adverbs with UVPRT?  Note that flagver=0; gainuse=0 gets you the 
same as maxver('') with less typing.

Eric