Results after flat compensation

["Gutzwiller, Michael" <mgutzwiller@lanvision.com>]

I've been working hard to get the flat compensation program tested and
characterized.  The results look very encouraging.  Here's what I've
been doing.

*	I took the 5 nights I've used before and flat compensated all
the star lists.
*	I then read in all the observations into my own little database
for testing.
*	I then created a catalog based only on good, photometric lists
(background variation < 0.007).
*	I then selected only those catalog stars seen in four or more
images in both V and I.

	Here are the results.  1281 stars passed all the tests.  The
variations (one sigma) of the magnitudes are listed below:

	Magnitude	Variation
		V	I	
	<7	0.024	0.070	
	7.0-7.5	0.014	0.037	
	7.5-8.0	0.016	0.039	
	8.0-8.5	0.015	0.040	
	8.5-9.0	0.017	0.045	
	9.0-9.5	0.025	0.042	
	9.5-10.0	0.022	0.050	
	10.0-10.5	0.034	0.057	
	10.5-11.0	0.045	0.084	
	11.0-11.5	0.069	0.118	
	11.5-12.0	0.106	0.153	
	12.0-12.5	0.128	0.167	
	12.5-13.0	0.194	0.149	
	>13	0.181	NA	
	
The variation in magnitude bottoms out in V at about 0.015 which is
about the error I would expect purely from the adjustments to the Tycho
catalog stars.  The variation in I bottoms out at about 0.040 which is
probably due to several factors:
*	The I Tycho magnitudes used to compensate the lists are
calculated from Bt - Vt.
*	The compensation was very large for one of my I cameras but not
the other.  Camera 0 had a compensation range of 0.07 magnitudes while
camera 2 had a range of 0.37 magnitudes so getting a resultant error of
0.04 magnitudes is great from my perspective!
*	There is some evidence of a small color difference between
camera 0 and camera 2.

Mike G.