GlueX Offline Meeting, April 3, 2013
GlueX Offline Software Meeting
Wednesday, April 3, 2013
1:30 pm EDT
JLab: CEBAF Center F326/327
- Review of minutes from the last meeting: all
- RF Bunch Selection ( ANALYSIS Library Factories): Paul M.
- Changes to BCAL Reconstruction: Will L.
- BCAL reconstruction test with pi0: Beni
- Update on Geant4
- Action Item Review
- Review of recent repository activity: all
ReadyTalk Desktop Sharing
You can view the computer desktop in the meeting room at JLab via the web.
- Go to http://esnet.readytalk.com
- In the "join a meeting" box enter the Hall D code: 3421244
- Fill in the participant registration form.
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#
Talks can be deposited in the directory
/group/halld/www/halldweb/html/talks/2013-2Q on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2013-2Q/ .
- CMU: Will Levine, Curtis Meyer, Paul Mattione
- JLab: Mark Ito (chair), Simon Taylor, Elliott Wolin, Beni Zihlmann
- Northwestern: Sean Dobbs
- IU: Matt Shepherd
- Mark announced that we will using the ReadyTalk desktop sharing method for these meetings in the future on a trial basis. See the instructions above.
- Elliott announced that for the online, since there is no previously existing code base, will go with the relatively new C++ 11 standard. This standard is 100% backward compatible with legacy C++ code. We briefly discussed whether the offline system should embrace this standard as well, though without any clear conclusions.
RF Bunch Selection
Paul described recent work on the RF bunch selection problem. Often ghost tracks and showers influence the selection. In some contexts he gets the wrong bucket about 20% of the time. Accidentals can effect the result as well. He has tried a method where a particular event topology is requested first. Only after an event satisfies the topology requirement is bunch selection attempted and then only using tracks and showers that are part of the given topology. He then gets a much higher success rate. He led us through a wiki page that describes the classes that implement these ideas.
We discussed the way the start time for tracking is done now. Each track determines a T0 for itself independently, using either the time-of-flight or the start counter, so there is no requirement for a pre-defined global start time to get the tracking going.
BCAL Reconstruction Work
Will has been updating the BCAL reconstruction to accommodate the new hit generation and smearing scheme implemented by David Lawrence. He showed two diagrams that outline the flow for both the old and new schemes. The new scheme should give more a more realistic representation of the hits, but also requires a time-walk correction (as will the real data). Will also remarked on the complication of having to accommodate the change in both the old KLOE code (the current default), and the new reconstruction scheme from Matt Shepherd.
Development is done. What is left is to do a comparison of results using the old and new simulated data schemes. Will will work on that.
Another Alternate BCAL Reconstruction Algorithm
Beni presented results from a new algorithm for BCAL reconstruction, independent of the two mentioned in Will's talk. He was motivated by a desire to understand the perennial problems we have seen with photon reconstruction and wanted to work with a reconstruction code that he understood well. He showed a simulation with 1 GeV π0s thrown at 50 degrees in polar angle and results from all three reconstruction codes. Overall, both the Matt code and the Beni do better than KLOE. Matt's appears a bit more mature at this stage. The KLOE code seems to be particularly bad for symmetric decays. More studies are planned.