You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Current »

Agenda

  1. Tracking                                       – Omar
  2. Software Round Table                   – All

Notes:

Omar updated us on plans for tracking. Notes taken during the meeting:

  • Most important task will be to really understand the tracking efficiencies. This will not require software changes (yet?), just more plots.
  • For vertexing, the resolution is not constant vs. z. (Sho has pointed this out at the collaboration meeting.) Norman is looking into this. Omar will create a JIRA task.
  • Omar suggests we should look at lower level details of the tracking as well, but there are no obvious issues at this point.
  • In the future, an important aspect will be to look at tracking performance (CPU).
  • There are older tasks on JIRA related to GBL that were never competed.

Questions:

  • Rafo asked about the tracking state at SVT layers. 
  • Rafo asked about the truth information in MC for mixed MC files (i.e. WAB-beam-tri). He wants to use this information for ECAL only analysis of MC and data.
    • It turns out that this is more difficult to keep around than at first glance. In G4 one would need to keep the entire inheritance tree of the particle shower to trace back to the original particle that came from the target. This would be a large thing to store. Then the way we do the combining of events from different generators further complicates the matter. It may be too big a task for too little gain to change all that.
    • Holly suggested that you can look at the scoring plane before the ECAL and correlate ECAL clusters with particle in the scoring plane. That cuts out most of the particle tree and makes it easier to get back to the original particles. 

Updates:

  • Sebouh gave us an update on the status of his re-write of the ECAL drivers. It is mostly completed. The suggestion was made to have Jeremy look at the code and the run the unit tests. Sebouh will check the code into the trunk after these are done. He will give a brief summary presentation of his work at a later meeting.
  • Holly has done work on implementing a timing correction to get the trigger associated with an RF bucket. This should improve overall timing. The result will be stored in as a delta-T in a new LC Generic object. A suggestion was made to store the original cluster time and location of the cluster that was used for the correction (or some other unique identifier of the cluster) so that analysts can figure out which cluster or hit was used.
  • Holly is also looking into how to smear the MC to make it closer to the data.

 

 

  • No labels