[daip] [!7243]: AIPS - UVPRT and scaling of visibility weights

[Michael Bietenholz]

Michael Bietenholz updated #7243
--------------------------------

              Status: Open (was: Closed)

UVPRT and scaling of visibility weights
---------------------------------------

           Ticket ID: 7243
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/7243
           Full Name: Michael Bietenholz
               Email: mbieten at yorku.ca
             Creator: User
          Department: AIPS Data Reduction
       Staff (Owner): Eric Greisen
                Type: Issue
              Status: Open
            Priority: Default
      Template Group: Default
             Created: 10 September 2015 08:13 PM
             Updated: 16 September 2015 03:32 PM



I'm still confused about the weights printed out by UVPRT

I ran UVPRT with DPARM=-1,0 (same little sample uv file I sent the other day),
antenn 3,0; basel 10,0; BIF 5; CHANNE 2 and got:

Weights have been multiplied by  1.0E+06
 ....
   Time      Ant   U(Kilo)  V(Kilo)    Amp Phas Wt    
  1/00:00:01  3-10  -32178.  -15852.   0.018 -53 48  

So the of this vis is 48e-6;  LISTR OPTY=MATX concurs.

Then I ran UVPRT w/ DPARM 0 and got
Weights have been multiplied by  1.0E+01
  1/00:00:01  3-10  -32178.  -15852.   0.018 -53  1  

So now it suggests that the weight is in fact ~0.1

I think that the true weight is ~48e-6, but in that case the DPARM 0 output should have rounded to 0, not 1.

Is UVPRT actually clipping the weights to the available range, and prints out "1" whenever the weight is +ive but less than the available range?  If so it should probably alert the user that a weight of "1" is just a flag indicating the true weight
could be anything lower than "1", since a straightforward interpretation of the DPARM=0 output would lead you to think that the weight is 2000 times bigger than it really is!  


------------------------------------------------------
Staff CP:  https://help.nrao.edu/staff