Mark IV pipeline, version 0.2

[Stupendous Man <richmond@a188-l009.rit.edu>]

  I've improved the pipeline in a few small ways.  

  First, one now has
the ability to specify two reference catalogs -- one for astrometry,
and one for photometry -- instead of using just one for both purposes.
One can still use the same catalog for both if one wishes.  With two
reference catalogs, we should be able to find decent astrometric
solutions even in very star-poor regions of the sky (by picking one
of the big Tycho2 subsets), yet not give up on the photometric accuracy.

  Second, after seeing how long (over 20 minutes!) it took my machine 
to go through the big astrometric reference catalog and pick out 
the stars for any given night, I decided to speed things up.  The
pipeline is written in TCL, not the fastest language.  I translated a
copy of the "refcat.tcl" program into Perl, which runs about four
times faster on my computer.  The pipeline will now try first
to run the Perl version of this step; if it fails (perhaps because
Perl isn't installed on the host computer), it then tries the 
good old TCL version.

  As one can see by looking at the "version.param" file, the latest
release is version 0.2.  The version number of the software is written
to one of the main output files when the pipeline runs, which may
help us to track down errors.   

  You can download this version at the usual place:

         http://spiff.rit.edu/tass/pipeline/

with a gzip'ed tarfile in

         http://spiff.rit.edu/tass/pipeline/pipeline.tar.gz

  Questions?  Comments?  

                                       Michael Richmond