Notes from preliminary SB meeting 13-March 2007
- Regular items
In view of the short meeting, this point was skipped.
- Review preparation
John reported on contacts he had with N.Young, who is actively contacting people to get users’ input for the review. The target was that input from users and experiments is being communicated to Geant4 by mid March. According to informal contacts, the following topics were raised:
- CPU performance. There is a concern of a degradation of CPU performance in HEP calorimeter applications.
- ‘FLUKA’ physics in Geant4; there were no further details on this vague request. Release notes; there were comments that teh release notes are not adequately describing the changes introduced. Also, sometime the language was criticized. It is expected that the more formal input from users will contain more detailed information.
The preparation of materials for the reviewers is in progress. For the
hadronic physics performance, a draft web site is under development and
can be accessed from http://cern.ch/geant4/results/validation_plots.
For standard electromagnetic physics, draft web pages are being prepared and feedback on these was requested. The low energy electromagnetic physics group is working on validation papers and advanced examples. In total 6 papers are being written, 4 of these are in an advanced state.
Action 1 on John Apostolakis: Request the questions from reviewers
Action 2 on all Steering Board members: Send questions for clarification on user input to both C. Young and John Apostolakis.
Adressing the CPU performance feedback has started. A set of 5 tests has been selected: a test of geometry navigation with the honeycomb calorimeter, two simple EM calorimeter on standard EM performance, and the simplified calorimeter using settings mimicking the CMS HCAL (50 GeV pi-, with two cases: 4T and 0 field). These tests have been ported back to a selection Geant4 versions as far as 5.2; checking performance of releases since then is in progress.It is planned for the future to keep these tests under system testing monitoring the Geant4 performance. Note that a set of benchmarks tests had been used in system testing in 2003-4, but porting and running were not undertaken for more recent releases.
Action 3.on John Apostolakis and Aatos Heikkinen: Contact Tatsumi K. and Alex H. on a test for monitoring performance for pysics including neutron HP
Planning of practice talks:
- 23-March: low energy electromagnetic physics
- 26-March: visualization
- 27-March: hadronic physics
- 28-March: standard electromagnetic physics
- Fermilab proposal to join Geant4
John had asked Marc Verderi and Vladimir Ivantchenko to prepare a draft recommendation letter. The discussion with both hadronic working group coordinators is ongoing. Input from two SB members was received, more input is welcome. In general, the proposal is seen positive, with a wish to extend the scope of the participation.
- Development items for 2007
The proposed plan of developments (http://cern.ch/geant4/support/planned_features.shtml) was distributed a week before the meeting. This plan of developments should be checked by working group cordinators such that is can be finalized in the coming weeks, see action item below. This proposed plan also includes a proposal for relases this year. The release end of June will be major release due to teh inclusion of changes affecting user code. The two main items afftecting user code are the introduction of tunable tolerances and the new parallel navigator. There could be an additional 8.3 release before 9.0 in May including corrections for stability problems seen by ATLAS and the modified FTF model.
Action 4. on Category Coordinators: Check the document for your
working group development items and give feedback to Gabriele.
Action 5. on physics lists task force: to provide work plans, including development items, by the next SB meeting.
- Collaboration Milestones
Regarding the next planned collaboration wide publication, Maria Grazia Pia confirmed she will provide a skeleton for the paper on physics configurations and physics lists, and gave a new timeline after the review.
The second project, the Geant4 physics book, will restart also after the review. A mini workshop on the Physics Book was proposed to help the project to make progress again.
Cross working group projects address the review and already known topics of user concern. The physics results web pages have to be checked ; old obsolete material should be removed, consistency with new draft pages (see obove) needs to be checked. A task force monitoring CPU performance has started to monitor performance of current and past G4 releases for typical setups, see above. The improvement of release notes needs more effort. Bug fixes should give a reference to teh bug, implications of changes should be made more explicit. A task force shall be formed to check release notes from a users point of view.
- Other formal items
New collaborators proposed:
- Andreas Schaelicke (DESY/Zeuthen) as individual proposed by and to join the standard EM working group. Per standard procedure, unless SB memebers not present raise new objections within two weeks, he be accepted as a candidate collaboration member. As he has already completed delivery of several processes he will be added to the public list of collaboration members immediately.
- Enrico Billi (INFN Bologna) as member of INFN proposed by low energy EM. He will become a PhD student later this year.
Previously proposed members:
- Ben Morgan (Warwich University/PPARC) as member of PPARC proposed by and to join software management working group. He had been propsoed by mail in October 2006, and has recently started contributing. He will now appear in the list of collaborators.
Tuesday, March 13, 2007 - 12:00