[evlatests] Referenced Pointing, with Durations

[Rick Perley]

    Following the successful recovery of referenced pointing with 
stop-time files, we tested to see if the procedure works with files in 
Durations. 

    Answer:      It doesn't. 

    Details:

    Observations were made of 3C273, at Kband.  Following an X-band 
referenced pointing determination, four short scans followed -- two with 
the offset solution applied, two with it not, alternating.  This cycle 
(including the reference pointing determination) was repeated once. 
    Ken confirms that offset solutions were made.
    The subsequent amplitudes show no difference whatever between the 
'on' and 'off' states. 

    Until this bug is squashed, high-frequency observations (especially 
in daytime) should be made using stop times.