Project

General

Profile

Geant4 EM Physics Code Review Weekly Phone Meeting

Meeting

Date: March 14, 2014 at 9.15 AM US CST (Fermilab local time)
Place: 1-866-740-1260 (ReadyTalk line) (Host : Krzysztof)

Participants:

John Apostolakis
Krzysztof Genser
Boyana Norris
Soon Yung Jun

Discussion

(Summary by Krzysztof)

Boyana walked us through the TAU data she recently collected and has uploaded to the wiki:

https://cdcvs.fnal.gov/redmine/attachments/16416/Lightweight_methods.txt
https://cdcvs.fnal.gov/redmine/attachments/16414/L2_data_misses_10.00.p01.txt
https://cdcvs.fnal.gov/redmine/attachments/16415/L2_inst_misses_10.00.p01.txt
https://cdcvs.fnal.gov/redmine/attachments/16417/res_stl.txt
https://cdcvs.fnal.gov/redmine/attachments/16418/br_msp.txt

The data are based on SimplifiedCalo, Standard run, 500 events with 50GeV e-, no magnetic field

The data look both useful and promising. Seeing to confirm some of our previous hypotheses.
Our initial comments suggestions:

Make the lists longer to allow to see more EM classes/functions.
Base the threshold on the % of time or instructions, not on absolute values.
Indicate which derived class caused invocation of a specific virtual function.
Possibly list the function signature (arguments).

We shall have our next meeting in a week, as usually.
We should check in as much of the report text as possible by then.