HDGeant4 Meeting, June 16, 2020

From GlueXWiki
Jump to: navigation, search

HDGeant4 Meeting
Tuesday, June 16, 2020
3:00 pm EDT
BlueJeans: 968 592 007

Agenda

  1. Review of minutes from the last Meeting (all)
  2. Issues on GitHub
  3. Pull Requests on GitHub
  4. Action Item Review
    1. preparations to advance to latest Geant4 release G4.10.06
    2. 20% reduction in cpu time from recent fix to tracking
    3. investigation of the multi-threaded bcal photon gun anomaly
    4. investigation of the backward kaon efficiency step at 120 degrees
    5. new HDGeant4 branch "bcal_nolight_heavy_rtj" for suppressing scintillation by heavy particles in the bcal
    6. yet to come - investigation of the 12% drop in preselection efficiency with CUTS card seen by George in etaprime,p

Minutes

Present: Sean Dobbs, Sergey Furletov, Colin Gleason, Mark Ito (chair), Igal Jaegle, Naomi Jarvis, Richard Jones, Susan Schadmand, Simon Taylor, Nilanga Wickramaarachchi, Beni Zihlmann

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

Review of minutes from the last Meeting

We went over the minutes from June 2.

  • Issue #152: BCAL Photon Gun Updates. This issue has been closed.
  • Issue #149: BCAL structure in E/p distributions for kaons. This issue has been closed.
  • Issue #143: BCAL reconstructed energy sometimes broken on startup: multicore only. Richard has been able to reproduce the problem of significantly different BCAL energy distributions on repeated runs of HDGeant4 with a 2 GeV photon particle gun at 60 degrees. This appears to only happen when hdgeant4 is run in multi-threaded mode. We generally build hdgeant4 against a multi-threaded build of Geant4, but even so, single-threaded mode for hdgeant4 seems not to be a problem. Richard noted that our running on the OSG is always single-threaded. Now that he can reproduced the problem, he will turn to its solution.
  • Issue #111: Difference in Acceptance between G3 and G4. Richard has produced a branch, bcal_nolight_heavy_rtj, where anything heavier than a proton will not produce visible energy. Colin will repeat the studies he presented last time with this branch and compare.
  • Issue #93: Calorimeter timing mismatch between g3 and g4. This issue still needs to be closed and linked to Issue #111.

Richard's Action Item Review

  1. Richard encouraged us to get started on preparations to advance to latest Geant4 release G4.10.06. Mark will build a version set, based on this release, and let Sean et al. know so they can do some comparative studies before we unleash the new release on the collaboration.
  2. Richard has discovered 20% reduction in cpu time from killing particles that go beyond the walls, floor, and ceiling of the Hall in the simulation.
  3. Investigation of the multi-threaded bcal photon gun anomaly, discussed above.
  4. Investigation of the backward kaon efficiency step at 120 degrees. There has been some progress on halld_recon Issue #378. See the issue for details.
  5. New HDGeant4 branch "bcal_nolight_heavy_rtj" for suppressing scintillation by heavy particles in the bcal, discussed above.
  6. "Yet to come": investigation of the 12% drop in preselection efficiency with CUTS card seen by George in etaprime,p

Action Item Review

  1. Build a version set with Geant4 10.06.04.
  2. Merge issue #93 with #111.