[daip] wideband calibration

[Walter Brisken]

Today I came to a clear realization that when calibrating my VLBA data 
taken at 4 widely separated IFs within L-band it is essential to produce a 
different model per IF to attain the best calibration.  What I would like 
to do is make and use separate images at each IF during each phase of 
calibration.  Ideally this would be done by making an image cube where the 
IF determines the frequency axis but I believe this is not possible within 
AIPS for various reasons.  A reasonable alternative would be to manually 
make images at each IF and determine calibration one IF at a time.  For 
example set BIF and EIF for each IF in turn and place values into the 
appropriate slots in SN or BP tables with each pass (keep adding values to 
the same table with each pass).  If this were possible then all downstream 
processing could be done in the standard way once the calibration is 
determined.  I notice that BPASS takes BIF and EIF, but based on the 
explain file it isn't clear it does what I need it to.  Neither CALIB 
nor FRING accept BIF/EIF.  The only way I can see to proceed with the 
current feature set is is by SPLATting out each IF before astronomical 
calibration is started, which is not ideal for bookkeeping and other 
reasons.

So basically I'm wondering if it is realistic to request BIF/EIF 
functionality in BPASS, CALIB and FRING that does what Ineed.

Thanks,

Walter