[evlatests] Summary of Modcomp-free performance

[Barry Clark]

In addition to what Ken said:

There were minor problems starting up - the operators eventually rebooted
the correlator controler, and things started working after (not necessarily
because of) that.

Flagging on VLA antennas off source is currently turned off, because the
flags would be always set and very annoying when the Modcomps are running
the array.  Possible cures:  1.) turn them on, but conceal them from the
operators.  2.)  Have the executor turn them on when it takes control, 
and provide a button on the operator screen to turn them off again.  (A
problem for the next couple of months at most.)

There is a bug in the setting of tolerance as a function of band by the 
Executor.

The really large delay corrections were 
X  EA13  B  10
X  EA14  B  22
X  EA21  B  17
X  EA14  C  13
X  VA04  D  10
X  VA10  D  60
X  EA21  D  26
C  EA17  A  40
C  VA28  A  13
C  EA17  B  32
C  EA17  C  46
C  VA06  D  18
C  VA10  D  20
C  EA17  D  35
L  EA21  A  25
L  EA21  C  19
L  EA21  D  16
K  EA17  A  61
K  EA21  A  18
K  VA28  A  14
K  EA17  B  56
K  EA21  B  18
K  EA21  C  23
K  EA14  D  27
K  EA17  D  60
K  EA21  D  20
Q  VA28  A  11
Q  VA10  D  21
Q  EA14  D  24
U  VA10  D  19

These do not coincide very well with Ken's list of weak antennas.  I have 
no guess why not.

I'll try to get them into the database in the next few days (between 
meetings, I guess).  X band is already entered.