Investigate physics setup differences between CMSSW and Celeritas

[whokion]

This is still a draft (WIP)

This summarize differences in simulation configurations between CMSSW and Celeritas and a list of tasks how to deal with them.

1. Propagation in the magnetic field

Field, Stepper and Region

	CMSSW	Celeritas
Field Type	DD4HEP (Volume-Based)	RZMapField
Stepper	DormandPrince	DormandPrince RK5(4)7M
RmaxTracker	8000 mm	3000 mm
ZmaxTracker	11000 mm	6000 mm

• [ ] Evaluate performance without the magnetic field (baseline): CMSSW vs. CMSSW+Celeritas.
• [ ] Add the along step action with the parameterized field in the tracker and evaluate performance.

Field Parameters

	CMSSW Default	Tracker	Vacuum	Celeritas
EnergyTh		0.2 GeV	0.015 GeV	N/A
MinStep	0.1 mm			1e-5 mm
MaxStep	150 cm		50cm
MaximumEpsilonStep	1e-02
MinimumEpsilonStep	1e-05			1e-05
DeltaOneStep	0.001 mm	1e-4 mm	0.1 mm
DeltaChord	0.002 mm	0.001 mm	0.1 mm	0.25 mm
DeltaIntersection	0.0001 mm	1e-6 mm	0.01 mm	0.0001 mm
MaximumLoopCounts	1000			100

• [ ] Add the energy threshold cut in the along step action.
• [ ] Evaluate the performance differences with different field parameters with the celer-g4 application.

2. EM Physics and Parameters

Physics list and constructor

	CMSSW	Celeritas
Default Physics List	FTFP_BERT_EMM	FTFP_BERT
EM Constuctor	CMSEmStandard_emm	G4EmStandardPhysics

EMM physics options and MSC paramters

	CMSSW	HCal/HGCal	Celeritas
DefaultCutValue	1.0 cm
G4GammaGeneralProcess	True		False
G4ElectronGeneralProcess	False		False
G4TrackingCut	0.025 MeV		from_import
G4MscRangeFactor	0.2	0.04	0.04
G4MscGeomFactor*	2.5
G4MscSafetyFactor	0.6		0.6
G4MscLambdaLimit	1.0 mm		1.0 mm
G4MscStepLimitType	Minimal	UseSafty	UseSafty

• Used only for the "Distance to Boundary" algorithm in G4UrbanMscModel (so, irrelevant)

• [ ] Add the Minimal boundary algorithm in celeritas::UrbanMscStepLimit.
• [x] Add the tracking cut in the along step action (imported from geant4).

Gamma processes and models

	CMSSW	Celeritas
Gamma General	Used	Not used
	G4PhotoElectric / G4PEEffectFluoModel	PhotonElectric / LivermorePEModel
	G4ComptonScattering / G4KleinNishina	Compton / KleinNishina
	GammaConversion / G4PairProductionRelModel	GammaConversion / BetheHeitlerModel
		(Raleigh / LivermoreRayleighModel)

e-/e+ processes and models

	CMSSW	Celeritas
	G4eIonisation / G4MollerBhabhaModel	EIonization /MollerBhabha
	G4eBremsstrahlung / G4SeltzerBergerModel+G4eBremsstrahlungRelModel	Bremsstrahlung / SeltzerBergerModel+RelativisticBremModel
Default	G4eMultipleScattering / G4UrbanMscModel+G4WentzelVIModel	msc / UrbanMsc
HCal/HGCal	G4eMultipleScattering / G4UrbanMscModel	msc / UrbanMsc
e+	G4eplusAnnihilation/G4eeToTwoGammaModel	EPlusAnnihilation/EPlusGG

• [ ] Test MSC without the G4WentzelVIModel in CMSSW.
• [ ] Add WentzelVIModel , the high energy MSC model (to do by @amandalund ).

[amandalund]

Great summary @whokion! I think we can check off the tracking cut which we should already support in Celeritas (#648 let us kill tracks with energies below LowestElectronEnergy in along-step and #709 added the ApplyCuts option).