[evlatests] Initial Results -- 2 GHz-wide data.

[Michael Rupen]

>
>    The subbands are misordered.  After some headscratching, the
> perturbation was clear.  Following is a table showing the subband order
> as presented by AIPS, and the order they should have been in:
>
>    Listed      Actual
>    --------------------
>       1               4
>       2              12
>       3                1
>       4                2
>       5                3
>       6                5
>       7                6
>       8                7
>       9                8
>      10               9
>      11              10
>      12              11
>      13              13
>      14              14
>      15              15
>      16              16
> ----------------------------
>
>    Another way to think about this is that the first two listed
> subbands are misplaced -- the first one belongs in slot 4, and the 2nd
> one belongs in slot 12.  All others are correctly ordered.
>

This was my stupidity I'm afraid.  Currently we have to begin with a few
OSRO scans before moving on to RSRO, to give the system time to set
up before invoking all those subbands (this should NOT be a long-term 
feature!).  For this observation I was foolish and used the same
subbands for OSRO as for RSRO.  Those OSRO subbands wind up with
the lowest Spectral Window IDs, and hence are stored first in the
FITS file.

For the L band observations Vivek is looking at, I used OSRO at X band
before RSRO at L band.  Hopefully the ordering that resulted there is
more sensible.  Still a few bugs to work out, but overall this is
very encouraging.


As to subband 11 -- that corresponds to a single Baseline Board pair.
If I'm doing my associations right I believe that's b101-b-{1,0}.
I'll have a look.   I would expect the L-band data to have the same
problem.

                  Michael