GlueX Offline Meeting, September 21, 2011
GlueX Offline Software Meeting
Wednesday, September 21, 2011
1:30 pm EDT
JLab: CEBAF Center F326
- 1 Agenda
- 2 Communication Information
- 3 Slides
- 4 Minutes
- Review of minutes from the last meeting: all
- Reconstruction sub-group reports
- Compressed Data Format Update: Richard or Igor
- BCAL Reconstruction Update: Will
- Plug-in to extract single particles from multi-particle events (email) : David
- Tracking explained: Simon
- Resources: an idea for managing large files: Dmitry
- Collaboration Meeting Agenda
- Action Item Review
- Review of recent repository activity: all
To connect by telephone:
- US and Canada: (866)740-1260 (toll free)
- International: (303)248-0285 (toll call) or look up toll-free number at http://www.readytalk.com/intl
- enter access code followed by the # sign: 3421244#
- ESNet: 8542553
- Meeting URL: http://evo.caltech.edu/evoNext/koala.jnlp?meeting=292I2iMeMB929nDs9aD9
- Phone Bridge ID: 10 1789
Talks can be deposited in the directory
/group/halld/www/halldweb/html/talks/2011-3Q on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2011-3Q/ . You have to be a member of the "halld" Unix group to do this.
- CMU: Will Levine, Paul Mattione, Curtis Meyer
- IU: Ryan Mitchell, Kei Moriya, Matt Shepherd
- JLab: Hovanes Egiyan, Mark Ito (chair), David Lawrence, Yi Qiang, Dmitry Romanov, Elton Smith, Sascha Somov, Simon Taylor, Elliott Wolin
- Regina: Andrei Semenov
- FSU: Nathan Sparks
- Mark announced that, by default, we will not be using the phone anymore at these meetings. If someone does happen to have a need for a phone connection, just let Mark know in advance.
- Elliott announced that he and David will be attending the CLAS12 Service Oriented Architecture Workshop. Contact Elliott for more information.
Reconstruction sub-group reports
Matt gave the report.
- switch from 3-3-4 segmentation to 1-2-3-4 segmentation
- geometry specification not as flexible as originally thought
- Will is working on it
- timing offsets
- talking to Mihajlo: time is set by time of highest energy block
- x,y,z is not taken into account
- will take some work and attention: Matt will work on this
Simon gave the report.
- some problems have to do with the quality of track candidates
- found issues with starting angle guess for very forward going tracks
- fixed up angle in track candidate
- some other minor changes
- farming out fiducial cut study a possible issue
- note: design goal for forward direction: down to one degree
Paul gave the report.
- revised PID scheme: see his slides
- original slides presented at last Physics meeting
- changes based on feedback received on email list
- referred us to the slides for details
- Beni: HERMES scheme used log of prob ratios
- Beni will send around description of the scheme
- idea: instead of adding chi-squared's, try to weight chi-squared from different sources
- the way to do that is not clear
- errors may be correlated
- motivation: what if a single hit on a track if spurious: isn't that an out-sized effect on the sum of chi-squared's?
- some skepticism expressed about validity of this concern
- update on uncertainties
- see his wiki page for details
- showed plots of measured uncertainties in position and energy for the TOF as a function of both position and energy
BCAL Reconstruction Update
Will described his recent work. See his wiki page for details.
- working on the new 1-2-3-4 segmentation of the BCAL
- did some studies with π0's and resulting errors from kinematic fitting (problem wiht phi error)
- David remarked that there is a mechanism to vary thresholds in the code now, but that is probably not how it will be done in the end.
- Consensus was that we should retain the KLOE reconstruction scheme as the default for now.
Plug-in to extract single particles from multi-particle events
David described a new plug-in that recasts one multi-particle into one (or more) single particle event(s). The user selects the particle type of interest and only that particle appears in the output event. This allows single particle studies with spectra that match that found with a multi-particle event generator. See his slide for details.
Simon gave us a summary of the method used for charged track finding and fitting. See his slides for details. In addition to describing the process, he also showed studies of the quality of the results, and outlined issues yet to be resolved.
Collaboration Meeting Agenda
We reviewed the proposed agenda for the Offline Software session. No objections were raised.