Project

General

Profile

Geant4 EM Physics Code Review Weekly Phone Meeting

Meeting

Date: October 4, 2013 at 10.00 AM US CDT (Fermilab local time)
Place: 1-866-740-1260 (ReadyTalk line) (Host : Krzysztof)

Participants:

Andrea Dotti
Krzysztof Genser
Boyana Norris
Soon Yung Jun

Discussion

(Summary by Krzysztof)

We concluded that based on the rather limited feedback from the Geant4 Collaboration
meeting we shall continue the review as previously planned.

Give the ongoing code developments and ref07 becoming more and more obsolete,
we shall use tag ref09a as the new baseline release. Soon will put it in the repository.

We discussed Andrea's hypothesis regarding a possible template specialization
of the vector implementation for pointers in the context of the
std::vector<std::vector<G4double>*> to std::vector<std::vector<G4double> change

We decided on the following plan:

Boyana will continue the cache misses analysis regarding
the impact of the changes to the implementation of
G4PhysicsVector and G4Physics2DVector.

Andrea will look into G4UniversalFluctuations/SampleFluctuations
and into the use of the random numbers in particular

Soon will look at classes deriving from G4VContinuousDiscreteProcess
and decide which one to review next (probably taking into account the original
list of compute intensive functions).

Krzysztof will look some more at one of classes deriving from G4VEMProcess,
probably G4eplusAnnihilation.

We shall meet again on Friday (10am Central).