[daip] Problems with FITLD

[Eric Greisen]

Francesca D'Arcangelo wrote:
> The files I've been running that had the FITLD concatenation issue
> were for project BM256H, February 28th '08.

Unfortunately, BM256 is a very large experiment.  I downloaded the first 
set of data (about 6 Gigabytes) and ran FITLD concatenating all.  It 
found data out of order so I wrote a quicky test task to report the out 
of order points (TIORD).  It turns out that for that set of data, the 
fifth file was observed later than the 6th and 7th.  MSORT fixed the 
problem quickly and INDXR was happy.

Let me know if you encounter a data set that "will not sort" to INDXR's 
satisfaction.

I should warn you that DBCON does not renumber antennas - this can be 
very important when concatenating multi-source VLBI data sets.  You need 
to run MATCH first for these.  Some documentation erroneously claims 
that DBCON does the renumbering.

Eric Greisen