[G12] Minutes from 1/13/2010 meeting

[Craig Bookwalter]

Hi folks,
 Minutes from today's meeting are pasted below. They'll be posted to the
wiki whenever the servers fire back up.

File Edit Options Buffers Tools Help
Agenda:
- update on gflux
- update on what's left for cooking
- update on momentum corrections


Attending:
- Dennis, Diane, Carlos, Mike P, Craig, Valery, Paul, Sasha O., Mukesh,
Ben, Johann
- Diane tested gflux as built by scons at JLab--it seems to be running
correctly but produces no output. She is still trying to pin down Eugene
to sit with her and debug
- Sasha is caching files at JLab from skim 5 for use in gflux--the total
size of the data is about 200GB so we should be able to maintain this on
disk.
- Johann reports that he tested a few of the failed jobs interactively and
they worked fine--he'll resubmit all the ones that failed when the farm
comes back up and see what happens. The failed files amount to about 8% of
the total data.
- Craig contacted Heghine about momentum corrections and she seems to have
been spurred to action. She requested events with energy loss already
included but Craig did not know what was meant by this. According to
Diane, an eloss function is available during the fitting for momentum
corrections, and it should be linked to Eugene's eloss package with its
corrections for the g11 target. This function should be called prior to
performing fits for momentum corrections.
- Valery commented that it is also possible to use GSIM to do energy loss
corrections and that in using this method you get corrections for the
entire CLAS detector, not just the region around the target, as in
Eugene's package.
- Diane showed some physics results on her laptop, as the wiki was down.
- Craig reports that he screwed up the start counter efficiency
calculation--when the start counter is not hit, the times are *not* set to
-10000 as with the RF vtime--Craig thinks the status flag is the actual
answer. ST hits with a nonzero status flag are good. He is currently
re-running the efficiency calculation.
- Mike has 800 eta Dalitz events with 25% of the data--with the full
dataset g12 will have the world's largest sample of eta Dalitz decays
(3200 events vs 180 events). Mike might also have the world's best sample
of omega->pi0 e+e- decays as well.
- Craig is working on a ++- skim for FSU, but trying to reduce the pass1
filesize by a factor of two or more if possible. Paul suggests cutting out
events based on stvtime--Craig will look into this.
- Craig suggests that there be an action item list on the wiki. He'll add
this as it comes back up.