Difference between revisions of "Mixing Simulated Events"

Revision as of 14:42, 23 July 2015

Goal: Include beam background and detector noise into simulation by mixing a library of background events with physics events

Tasks

• Compare simulated beam background with data
  • Use PS triggered data
    • Nathan's skims: /volatile/halld/home/nsparks/ps_skim/data
    • Plan: Run 3185, use detcom 02 settings for simulations, which were calibrated to this run
  • Caveat: Simulations only include beam background, not detector noise
• Event merging mechanics
  • Update HDDM C++ API so that it can be used for hddm_merge_events
  • Implement hit merging after mixing
    • Do this in hddm_merge_events?
  • mixing with HDDM vs. EVIO formats
    • Data is in EVIO format
    • Could add class to write out events in hddm::s format, similar to danarest/DEventWriterREST
  • Do we only merge full events, or do we mix hits from several events?
• Event library management
  • how much disk space per event is used? what is the total disk footprint needed?
  • where should event library be stored?
  • Mixing random trigger events works when trying to simulate events with the conditions for a given run. What do we do when we want to simulate conditions for which we don't have data yet (e.g. high-intensity running)?
  • Keep reminding Somov to implement in- and out-of-time random triggers