Hi, this message should mostly concern Tancredi, Adrian and Nickolas.
Today when I left, show_single.C gave the ridiculous results of tof
efficiency larger than one. I think I have spot the problem and corrected
it.
The way the script work (also in cc_eff.C) is that it fill histograms of
whenever tof paddle 0, 1, 2, or 3 is hit, then get the number of events in
those histograms by GetEntries() function. However, when there are two
or more hits in those paddles (for instance pad 1 and 2 are both hit),
which takes a considerable fraction in all the events even with field on,
we count the event twice while in the proton histogram, the event is
counted only once. So the numerator is bigger than it should be.
The problem is fixed by adding to the cut the requirement that no more
than 1 paddle are fired. so we are looking only at the events with 1
proton at backward angle and 0 or 1 tof hit in the forward angle. See
show_single.C. It prints out three sets of numbers and the last set using:
#events(tof_hit in 0or1or2or3)AND(cc_hit in box0)AND(proton in pad 14)
-------------------------------------------------------------------------
#events(cc_hit in box0)AND(proton in pad 14)
should be close to the efficiency. The number I get is arround 90%. Of
course details must be checked.
On the other hand, cc_eff.C does not have this problem since the eep
ntuple is used there and the events included are 1 left 1 right.
I send this message just so we know the known bug with show_singles.C is
fixed, I am sure there are still bugs, lets keep working on it.
Cheers
Chi
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:28 EST