HDGeant4 Meeting, April 21, 2020
Tuesday, April 21, 2020
3:30 pm EDT
BlueJeans: 968 592 007
- Review of minutes from the last Meeting (all)
- Issues on GitHub
- Pull Requests on GitHub
- Action Item Review
Present: Mark Dalton, Sean Dobbs, Mark Ito (chair), Igal Jaegle, Naomi Jarvis, Richard Jones, Simon Taylor
There is a recording of this meeting on the BlueJeans site. Use your JLab credentials to get access.
Review of minutes from the last Meeting
We went over the minutes from March 24 without comment.
BCAL structure in E/p distributions for kaons
We went over Issue #149, recently submitted by Peter Pauli. He sees sharp features in the distribution of E/p in the BCAL for K−s in hdgeant4 and not in data. They are present both in the simulation of his reaction and in the single particle gun. The discussion went on for quite a while as we worked our way through the plots and comments in the issue. Please see the recording and the postings in the issue for all the details. In summary:
- We thought one way to untangle the various features that Peter has pointed out might be by turning on and off (a) decays and (b) hadronic interactions. Note that at Beni's suggestion, Peter has already looked at decays.
- Richard is suspicious that there might be a problem in the treatment of kaons stopping in the BCAL. If energy released by capture of stopped kaons is not accounted for in the Monte Carlo, then that could explain the difference Peter observes between MC and data. Richard thought that before Peter spends time on item(1), he (Richard) should take a look at this.
Reports from UConn
- Richard has some students working on MC. They have discovered some issues, including seg faults, that they have reported to Richard. At this point they have not been submitted as GitHub issues.
- Richard raised the issue of upgrading our version of Geant4.
- We are at 10.02.p02, released June 29, 2016. The latest release is 10.06.p01, released February 14, 2020. That is a big gap.
- About half of the custom fixes from Richard that we needed for 10.02.p02 are incorporated in 10.06.p01. He has produced the correspondingly reduced set of custom fixes need for this latest release, as well as fixes need for other releases since 10.02.p02.
- Going to 10.02.p02 requires GCC 5.x. We are using 4.8.5 on RHEL7 and CentOS7. Past experience tells us that this is no small deal.
- There is a start-up delay with G4 10.02.p02 that was not present in 10.02.p02 due to "voxelation". There is an option to skip this step, at a possible cost in execution speed.
- When working remotely, there is a package called VirtualGL that will allow graphics commands to be executed on the local graphics card. This speeds things up considerably.