[evlatests] 3-bit tests, trying various levels.

[Vivek Dhawan]

I have looked at 3 of these from yesterday. They are too hard to interpret in
detail. The lowest bits are not much worse than the higher ones, but there were
too many if, ands & buts.

I think we should re-do the tests. Some grudges against the data: Shadowing;
low elevation on 3C147, different shadowing on 3C147 vs. 3C84 (and nearby
blank field). Plus large delay changes on some antennas between the 3 files
(~200ns). Plus large bandpass changes on some antennas.

If these same files are run again, starting at 9 or 10pm instead of 6
would be great; and a reset of delays before letting them rip.

>> 3-bit tests.  Only antennas 6, 7, 10, 14, 18, 23, 26 are being used.
>>
>> 01Dec 00:25:25
>> Starting setBits-64.evla
>>
>> 01Dec 00:27:50
>> Starting ku3eq_3C84plus.evla
>>
>> 01Dec 00:31:00
>> Antennas are on source.
>>
>> 01Dec 01:21:30
>> Starting setBits-32.evla
>>
>> 01Dec 01:24:33
>> Starting ku3eq_3C84plus.evla
>>
>> 01Dec 02:18:00
>> Starting setBits-16.evla
>>
>> 01Dec 02:20:23
>> Starting ku3eq_3C84plus.evla