I found a bug in the monitor box decoding: box number was
set to 98 for ERP HW channel 40, but then after that
there was "if box>80 box=box-80", the standard ERP thinger for boxes
1-4: this was setting then the Bari box number to 18. This is easy
to fix.
Under normal circumstances, getting the Bari box number wrong should
cause few problems, because same-time events get vetoed regardless of
box number. (Well, actually, some Bari box hits wouldn't get vetoed
for the cases when there's a Bari trigger with no corresponding ERP
event, so in fact it is a good thing for the program to know that the
Bari box is box 98). However, for this run, the UT time fit at the
beginning of the run messed up, so that inter-SM coincidences were not
getting dealt with properly. I'm still working on why the UT time fit
failed -- it worked OK when I ran the monitor code on the run 10529
disk file, so possibly the problem is related to a missed buffer (?).
I'll have to work on this some more.
Kate.