search

GRIFFINCollaboration / detectorSimulations_v10

Geant4 version 10 of the simulation code for the GRIFFIN array and it's suite of ancillary detection systems.
MIT License
9 stars 14 forks source link

Geant4 Version Support #40

Open cbray0 opened 6 years ago

cbray0 commented 6 years ago

Feature support in different versions of Geant4:

Geant4 Version: 10.01 10.02 10.04
Angular Correlations: With G4GGAC10.01 - Native Support
Spectra Accuracy: Gamma de-excitation bug (Addressed in #35) Correct (AFAIK) Correct (AFAIK)
detectorSimulations_v10 support: Release 1 Master Unsupported/Incompatible

Are there any plans to update detectorSimulations_v10 to work with Geant4.10.04?

AndrewMac1 commented 6 years ago

For version 10.04 the Geant4 group has implemented their own angular correlations into the radioactive decay package. You can see it in the release notes:

https://geant4.web.cern.ch/geant4/support/ReleaseNotes4.10.4.html

You just have to enable them, so the hack made to the version that we used (10.01) will no longer be needed for later releases. I have not personally tested the functionality of this updated release but they say it is functional.

AndrewMac1 commented 6 years ago

To enable it you can use the following I believe.

https://geant4.web.cern.ch/geant4/UserDocumentation/UsersGuides/ForApplicationDeveloper/html/AllResources/Control/UIcommands/_process_deex_.html

cbray0 commented 6 years ago

I am aware that Geant4.10.04 has native support for angular correlations. However, as the detectorSimulations_v10 Readme states, detectorSimulations_v10 only supports Geant4.10.02, and will not build against Geant4.10.04. Are there any plans to support newer versions of Geant4 going forward?

AndrewMac1 commented 6 years ago

Ah sorry I misread this thinking it was only for the angular correlations patch. I do not believe anyone has started working on this yet for 10.04. Maybe @VinzenzBildstein has a better idea on this and if there is even a timeline for this yet.

VinzenzBildstein commented 6 years ago

There is no plan as of yet to update the code to run with geant4.10.4 (or geant4.10.3).

I think we should make the code work with the newer versions, someone will just have to do the work. And that would include not only changing detectorSimulations_v10 to compile with geant4.10.4, but then also verifying that the results are correct. The latter is the bigger job, I believe.

VinzenzBildstein commented 6 years ago

I've made a new branch geant4.10.04 to test the simulation compiled vs. geant4.10.04.p01. Apart from the changes to the particle iterator (was already necessary for geant4.10.03), I also had to comment out the explicit use of the (obsolete) G4FermiBreakUp model in PhysListHadron. I haven't checked if this affects the simulation.

The simulation ran about as fast as with geant4.10.02 (3-4 minutes), which is much faster than it ran with geant4.10.03 (140 minutes). I'm not sure if there was something else going on to slow that version down so much, but since geant4.10.04 seems to work I think we don't need to bother with that version anyways.

I will update the readme to mention the new branch (and that this branch hasn't been tested yet). If people could test their simulations with the new branch and report any problems or successes, please let me know.