Offline analysis of AIDA data files from R31_0 to c. R31_5 (elapsed time 1120.45s)

Note impact of increased noise ( https://elog.ph.ed.ac.uk/AIDA/494 )on LEC spectra,
particularly for DSSSDs #1 & #2.

*** ENTRY finish

 *** data items: 1231535778 ( 133914.33 Hz)
 *** ADC events:  112061604 (  12185.32 Hz)
 *** time warps:          0 (      0.00 Hz)
 *** DSSSD # 1 count:  24147717 old count:  24147349 dt:   1120.45 s  LEC rate:  21551.85 Hz
 *** DSSSD # 2 count:  22686292 old count:  22685929 dt:   1120.45 s  LEC rate:  20247.53 Hz
 *** DSSSD # 3 count:  13051178 old count:  13050922 dt:   1120.45 s  LEC rate:  11648.18 Hz
 *** DSSSD # 4 count:  11173273 old count:  11173038 dt:   1120.45 s  LEC rate:   9972.15 Hz
 *** DSSSD # 5 count:   9478275 old count:   9478090 dt:   1120.45 s  LEC rate:   8459.36 Hz
 *** DSSSD # 6 count:  15562532 old count:  15562263 dt:   1120.45 s  LEC rate:  13889.57 Hz
 *** DSSSD # 1 count:       165 old count:       165 dt:   1120.45 s  HEC rate:      0.15 Hz
 *** DSSSD # 2 count:       275 old count:       275 dt:   1120.45 s  HEC rate:      0.25 Hz
 *** DSSSD # 3 count:       523 old count:       523 dt:   1120.45 s  HEC rate:      0.47 Hz
 *** DSSSD # 4 count:       469 old count:       469 dt:   1120.45 s  HEC rate:      0.42 Hz
 *** DSSSD # 5 count:       297 old count:       297 dt:   1120.45 s  HEC rate:      0.27 Hz
 *** DSSSD # 6 count:       118 old count:       118 dt:   1120.45 s  HEC rate:      0.11 Hz

 *** ENTRY finish ends

S O R T    S T O P P E D ..... Tue Nov 29 14:18:07 2016

Spectrum s400 displays the frequency of bits 0-47 of the (MBS) 25MHz correlation scaler.

For those bits which can be expected to cycle often within the sampling period we would
expect to observe each bit with equal probability and this is observed for bits 0-31.
Conclude there is no evidence for hardware/firmware problems with the AIDA 25MHz correlation
clock for bits 0-31 and therefore probably not for bits 32-47 either.

I think the SyncCheck issues observed could due to:

1) the effect of sampling data from the AIDA TapeServer data buffers,
2) arithmetic/coding bug in the SyncCheck code
... or ?

I don't recall observing these issues before ...  does that mean we weren't looking? ... or
that there has been some change to hardware operation, software ... ?