- AM, DH: bgrid2.blast / bgrid.blast. See other emails. There was a bug
in the interpolated field-map data discovered during implementation to
blastmc v3. Bug did not affect recon (still using calculated map).
- blastmc: AM still is *the* librarian. Please contact him for changes
to the montecarlo
- pro2003: all online software can be started from the ~/pro2003
directory. All the sources are available as well. The
pro200XXX directory will become a full module in CVS
_ analysis: a comment from CC. REsults of last week (resolutions, recon)
did not use right wire.cal file !! Expect improvements. Next
results/crunch after v2 (end of week, report on wendsday).
Report status of recon code (wang)
- calib: shows effect new drift time tables (garfield). Stubs line up
much better with individual hits also at cell boundaries.
Shows differences < 1 mm
dh: may add slope parameter to each wch TDC channel to line up
stbs even furhter. New t0 offsets needed for new drift time
tables.
fast fit: presently all tracks are swimmed and the one with the
best X2 is selected. This is not a time-effective way to go.
In final v2 candidate tracks will be fit only in the Wch (no
swimming)
- cc: shows blastrc to source reconstruction options for reconstruction
general format is "program.class.option value" where
program=blastmc/nsed/recon class=TBLTrack, TBLNewt, TBLFieldCoil ...
is the class where the option is used and option could be a cut (max
chi2 allowed) or a true/false switch each options carries the name of
the class where it is used.
_ discussion about documentation: source code and user. TB: should be
minimal, people really have to read the code but a general introduction
(what is simTrack, fast fit) and a few hosto's (online) needed. See also
hauke request. Probably some hands on demonstration needed during startup
Note that with a distributed system like ours what you will do online is
90% of what you will need to do offline (but of course you'll have ample
choice of user functions and cuts..!). TB/MK are involved, while core
group (TL,WX,CC,CZ) progresses on c++ analyzer. MK: need at least
class documentation. Training will require active partecipation by users
who are not very familiar.
- Database: discussion about runlist table (AS). Runlist table will only
have start/stop/run# entry. Critical config files have own timestamped
table. Interface to epics hv also discussed. We seem to be converging.
________________________________________________________________
________________
Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124
research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST