CERN Accelerating science


Version 1.0

  • Geometry Biasing with Importance:
    • Understand the differences between generic biasing applied to geometrical biasing and the "standard" importance biasing
    • Improve testing of biasing processes - to identify and understand differences and compare with analogue simulations
    • Clean up and revise test33 to correctly test biasing in MT mode
    • Investigate the "switchability" between generic and geometrical biasing
    • Consider deprecating importance biasing
    • Debug the geometrical biasing for the case of deleted and reinstantiated geometries.
    • Command line and "smart" biasing scheme

  • Reverse Monte Carlo:
    • Improvments of EM ReverseMC for case of thick shielding
    • Continue/Finish Migration of ReverseMC to MT

  • Generic process:
    • Consider interest of new method G4VProcess::StartEvent()

  • Generic Biasing:
    • Continue enriching event biasing options:
      • Refactor existing generic biasing brem. splitting example to source
      • Leading particle (alternative to existing one in HAD)
    • Statistical test suite to verify correctness of biasing wrt to analog
    • Improvements:
      • Allow/demonstrate use of parallel worlds
    • Feasibility studies and prototyping
      • Biasing of charged particles, with cross-section changing over the step
      • Use of occurence biasing to allow continuous density change inside a same volume
      • DXTRAN-like biasing
      • Material/isotope biasing
      • Woodcock tracking
      • Implicit capture

  • Materials:
    • Evaluate usage of G4float for material data instead of G4double
    • Improved G4Exception usage in material classes, remove asserts
    • Introduce some c++11 into material classes
    • Evaluate interest of a material extension through an abstract class mechanism to at least gather information about processes using such extended properties (eg : DNA, channeling, thermal neutron)

Updated: 29th February 2016