Difference between revisions of "June 26, 2007 Calorimetry"

From GlueXWiki
Jump to: navigation, search
(Agenda)
(Minutes)
Line 21: Line 21:
  
 
== Minutes ==
 
== Minutes ==
 +
 +
Present:  Simon, Elliot, Elke, Alex, David, Eugene, Matt B., Paul, Blake, Zisis, Richard, Beni, Mihajlo, and Matt S.
 +
 +
* Kinematic Fitting (Matt B.):  Matt has been working with events in which &pi;<sup>0</sup>'s were produced to try to do mass constrained kinematic fits.  Much effort is being spent in developing the fitter in a general way that uses the objects in the reconstruction framework and that will be extensible to full event kinematic fits.
 +
 +
* Pythia (Eugene):  Eugene showed a table of various various exclusive final states and cross sections as measured and as generated by Pythia.  Pythia predicts the total cross section to be about a factor of 1.6 smaller than measured.  In general for events with higher multiplicity final states especially those with strangeness Pythia tended to overestimate the cross section.  Elke suggested that some of this could be tuned by adjusting various thing such as the strangeness suppression factor.  More work needs to be done to check whether or not Pythia is appropriately handling resonant substructure of multi-hadron final states and how baryon production is handled.  It was suggested that perhaps generating MC according to a compilation of various cross sections with some rough resonant structure might be better than using Pythia.  This will be investigated.
 +
 +
* ToF (Beni):  Beni reported on some work he has done to make the digitization more realistic for the ToF that includes time smearing from recent NIM publications and energy smearing consistent with scintillator response, photostatistics, and attenuation.  Beni's revisions also include changes to the ToF hit structure to pass along true hit locations of charged particles on the ToF wall.  This report provoked some discussion about how truth information is maintained in the simulation.  The original philosophy was to not propagate truth information along with detector response from HDGeant into DANA in the same structure.  However, since we have started writing a fair bit of the detector response and digitization in XXXResponse objections in DANA it has become clear that we do want to pass generator level info in this way.  One must be sure though that the output of the XXXResponse factories actually does mimic real raw data and contains no truth info that the reconstruction can use to cheat.  David suggested that this could be done by writing the response output to disk and stripping away the truth info before doing reconstruction.  As we further develop the simulation and digitization we will need to keep an eye on this.
 +
 +
* BCAL (Alex):  Alex reported that he was looking in detail at the cosmic ray runs taken in Hall B during the beam test.  Cosmic rays are particularly good for doing a cell-to-cell calibration since they are stable and the average path length in a cell is the same.  Also this calibration technique is independent of that utilized by Blake in his photon-run analysis.

Revision as of 16:56, 26 June 2007


Agenda

This meeting will be a collection of short reports on status of various projects

  • Eugene on Pythia background simulation[1]
  • Matt (Bellis) on kinematic fitting progress
  • Beni on ToF digitization in software Media:tofdigi1.pdf
  • Alex on cosmics analysis with BCAL
  • Zisis on BCAL beam analysis
  • review of progress: PID Progress
  • discussion of format for July workshop (Please Register)
  • Blake: Resolution vs Angle
Angleres1.gif


Dial-In Information

Minutes

Present: Simon, Elliot, Elke, Alex, David, Eugene, Matt B., Paul, Blake, Zisis, Richard, Beni, Mihajlo, and Matt S.

  • Kinematic Fitting (Matt B.): Matt has been working with events in which π0's were produced to try to do mass constrained kinematic fits. Much effort is being spent in developing the fitter in a general way that uses the objects in the reconstruction framework and that will be extensible to full event kinematic fits.
  • Pythia (Eugene): Eugene showed a table of various various exclusive final states and cross sections as measured and as generated by Pythia. Pythia predicts the total cross section to be about a factor of 1.6 smaller than measured. In general for events with higher multiplicity final states especially those with strangeness Pythia tended to overestimate the cross section. Elke suggested that some of this could be tuned by adjusting various thing such as the strangeness suppression factor. More work needs to be done to check whether or not Pythia is appropriately handling resonant substructure of multi-hadron final states and how baryon production is handled. It was suggested that perhaps generating MC according to a compilation of various cross sections with some rough resonant structure might be better than using Pythia. This will be investigated.
  • ToF (Beni): Beni reported on some work he has done to make the digitization more realistic for the ToF that includes time smearing from recent NIM publications and energy smearing consistent with scintillator response, photostatistics, and attenuation. Beni's revisions also include changes to the ToF hit structure to pass along true hit locations of charged particles on the ToF wall. This report provoked some discussion about how truth information is maintained in the simulation. The original philosophy was to not propagate truth information along with detector response from HDGeant into DANA in the same structure. However, since we have started writing a fair bit of the detector response and digitization in XXXResponse objections in DANA it has become clear that we do want to pass generator level info in this way. One must be sure though that the output of the XXXResponse factories actually does mimic real raw data and contains no truth info that the reconstruction can use to cheat. David suggested that this could be done by writing the response output to disk and stripping away the truth info before doing reconstruction. As we further develop the simulation and digitization we will need to keep an eye on this.
  • BCAL (Alex): Alex reported that he was looking in detail at the cosmic ray runs taken in Hall B during the beam test. Cosmic rays are particularly good for doing a cell-to-cell calibration since they are stable and the average path length in a cell is the same. Also this calibration technique is independent of that utilized by Blake in his photon-run analysis.