Difference between revisions of "HDGeant4 Meeting, July 20, 2018"

From GlueXWiki
Jump to: navigation, search
(first version)
 
(Agenda)
 
(One intermediate revision by one other user not shown)
Line 8: Line 8:
  
 
# [[HDGeant4_Meeting,_July_6,_2018#Notes|Review of notes from last time]]
 
# [[HDGeant4_Meeting,_July_6,_2018#Notes|Review of notes from last time]]
# Review of first round of simulation results.
+
# [https://halldweb.jlab.org/gluex_sim/G3G4_Comparisons/ver01/ Review of first round of simulation results.]
 
# [https://github.com/JeffersonLab/HDGeant4/issues Issues on GitHub] (all)
 
# [https://github.com/JeffersonLab/HDGeant4/issues Issues on GitHub] (all)
 
# Meeting Schedule
 
# Meeting Schedule
 +
 +
== Notes ==
 +
 +
Present:
 +
* ''' FSU: ''' Sean Dobbs
 +
* ''' JLab: ''' Alex Austregesilo, Thomas Britton, Mark Ito (chair), Justin Stevens, Simon Taylor, Beni Zihlmann
 +
* ''' Regina: ''' Zisis Papandreou
 +
* ''' UConn: ''' Richard Jones
 +
 +
There is a [https://bluejeans.com/s/PCm4D/ recording of this meeting] on the BlueJeans site. Use your JLab credentials.
 +
 +
=== Review of notes from last time ===
 +
 +
* [[HDGeant4_Meeting,_July_6,_2018#Notes|Notes]] were reviewed
 +
 +
=== Review of first round of simulation results ===
 +
 +
* Thomas showed [https://halldweb.jlab.org/gluex_sim/G3G4_Comparisons/ver01/ list of plots].
 +
** [https://mailman.jlab.org/pipermail/halld-offline/2018-July/003259.html email announcement of plots]
 +
* Much time spent discussion of one plot:
 +
 +
[[File:Prot 4 15 Hist Reconstruction.png]]
 +
 +
* Sean: geometry could be the cause
 +
* Richard: would like help isolating problem to simulation or to reconstruction
 +
** Thomas will put together a plug-in to explore
 +
* Thomas: both FDC and CDC hits are missing
 +
* question: are truth hits along real track trajectory
 +
** can be explored with event viewer
 +
** Beni volunteers
 +
* This is an [https://github.com/JeffersonLab/HDGeant4/issues/54 issue on GitHub]
 +
 +
=== Other Issues ===
 +
 +
* Maria Patsyuk [https://groups.google.com/forum/#!topic/gluex-software/VOnAjfCf3r4 reported overlaps in the geometry]
 +
** Richard will work on this
 +
* Richard: issues with roll-your-own event generators
 +
** interaction z and t are correlated by machine bunch structure
 +
** decay of long-lived particles can lead to double counting (parent and daughter particles both appear in final state)
 +
** [added in press] Richard wrote [https://halldweb.jlab.org/wiki/index.php/Guide_to_Monte_Carlo_event_timing_and_detached_vertices_in_HDGeant/4 new wiki page describing best practices].
 +
*** There is a [https://github.com/JeffersonLab/HDGeant4/issues/55 related issue on GitHub]
 +
* Sean added an additional plot on the [https://github.com/JeffersonLab/HDGeant4/issues/10 time shift in FCAL showers issue]. He sees a difference in hit times vs. shower times.
 +
 +
=== Meeting Time ===
 +
 +
No one objects to meeting on Tuesdays at 2 pm, every other week.

Latest revision as of 12:54, 21 July 2018

HDGeant4 Meeting
Friday, July 20, 2018
10:00 am
JLab: CEBAF Center, A110
BlueJeans: 968592007

Agenda

  1. Review of notes from last time
  2. Review of first round of simulation results.
  3. Issues on GitHub (all)
  4. Meeting Schedule

Notes

Present:

  • FSU: Sean Dobbs
  • JLab: Alex Austregesilo, Thomas Britton, Mark Ito (chair), Justin Stevens, Simon Taylor, Beni Zihlmann
  • Regina: Zisis Papandreou
  • UConn: Richard Jones

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

Review of notes from last time

Review of first round of simulation results

Prot 4 15 Hist Reconstruction.png

  • Sean: geometry could be the cause
  • Richard: would like help isolating problem to simulation or to reconstruction
    • Thomas will put together a plug-in to explore
  • Thomas: both FDC and CDC hits are missing
  • question: are truth hits along real track trajectory
    • can be explored with event viewer
    • Beni volunteers
  • This is an issue on GitHub

Other Issues

Meeting Time

No one objects to meeting on Tuesdays at 2 pm, every other week.