HDGeant4 Meeting, April 9, 2019

From GlueXWiki
Jump to: navigation, search

HDGeant4 Meeting
Tuesday, April 9, 2019
3:00 pm EDT
JLab: CEBAF Center, A110
BlueJeans: 968592007

Agenda

  1. Review of minutes from the last meeting
  2. Do-over on comparison studies
  3. Geant3/4 comparisons for the K-, K+ and p particle gun (Peter)
  4. Issues on GitHub (all)
  5. Pull Requests on GitHub (all)

Minutes

Present:

  • CMU: Naomi Jarvis
  • Glasgow: Peter Pauli
  • JLab: Alex Austregesilo, Thomas Britton, Mark Dalton, Stuart Fegan, Mark Ito (chair), Simon Taylor, Beni Zihlmann

There is a recording of this meeting on the BlueJeans site. Use your JLab credentials to access it.

Review of minutes from the last meeting

We went over the minutes from the March 26th meeting without significant comment.

Do-over on comparison studies

Simon is working on analysis of his particle gun simulations.

Geant3/4 comparisons for the K-, K+ and p particle gun

Peter led us though the set of summary slides from his particle gun studies. He chose protons and charged kaons because they are relevant to his analysis. Further he chose kinematic regions for each that match his reaction. He created monitoring plots for each species and had posted the complete set of plots for each as part of his GlueX Note.

The summary alone runs to 72 slides and we went through them all in the course of about an hour. The secretary does not have detailed notes on each slide, but will record here some general remarks:

  • Peter told us he chose plots that seemed interesting, either because they showed good agreement between GEANT and Geant4 or, more likely, because there were puzzling differences.
  • As a general guide, if two plots are shown on the same pad, the blue is GEANT and the red is Geant4. If there are two separate pads, then the top is GEANT, the bottom is Geant4.
  • He sees a charged particle reconstruction efficiency for Geant4 that is about 10% lower than that for GEANT. Alex noted that this is similar to what he is seeing when he studies ρ0 events.
  • Peter mentioned that the differences he picked out are mostly of two broad classes: (1) energy or (2) timing, in each case as measured by the calorimeters.
  • We noted that since the z-position of charged particles is not well measured for single particles, some the the timing distributions may be hard to understand. Still they should have similar features in GEANT vs. Geant4.
  • The versions of the code he used came before Richard's most recent pull request, symmetrizing the treatment of timing between the FCAL and BCAL. Some of the differences may have lessened in the new code; this should be checked.

Please consult the recording if interested in a plot-by-plot discussion.

Issue on GitHub: missing thrown events

Thomas called our attention to Issue #101 on GitHub, "Differences between G3 and G4 in eta.pi," submitted by Carlos Salgado. Carlos sees a completely different thrown momentum distribution for protons when using GEANT vs. Geant4 even though the generated events are identical in the two cases. Indeed they come from the same HDDM file.