Simulation Project Leaders Meeting 4-November-2010 ==== Attendees John Apostolakis (JA), Andrea Dotti (AD), Witold Pokorski (WP), Gabriele Cosmo (GC). Excused: Pere Mato, Alfredo Ferrari, Rob Veenhof. --------------------------------------------------------------------- ==== Minutes - Nothing to report. --------------------------------------------------------------------- ==== Simulation physics validation & framework - Andrea - Validation Studies: o Mainly concentrated in validating the development releases (Geant4 monthly reference tags) with simplified calorimeter, in view of the upcoming public release of Geant4 in December. The October development release is considered as the baseline, at least for what concerns the more stable and used models, namely QGS, BERT and FTF. o The activity for the validation of diffraction (fourth benchmark) has restarted. The software used in 2007 has been re-enabled and data from HELIOS and, possibly NA22, for target diffraction experiments have been studied. - Meetings: o The last LCG Physics Validation Meeting has been held on September 22nd with a contribution from ATLAS EndCap Calorimeter showing the validation of Geant4 9.3.p01 release compared to 9.2.p03. This release showed that the Geant4 simulation is generally improved in describing test-beam data. The considered calorimetric observables are: pion resolution and response, fraction of energy in longitudinal layers. The QGSP_BERT physics-list is very stable and gives basically the same results across releases; FTFP_BERT shows a clear improvement in describing the data; CHIPS shows a poor description of the data (should be noted that the version used for the comparison was still under tuning). These conclusions are in agreement with what has been observed by other validation studies (ATLAS HAD barrel calorimeter and CMS calorimeters). --------------------------------------------------------------------- ==== Generator services - Witek - Meetings: o Presentation about the status of GENSER at that CHEP10 conference. o The next meeting of GENSER will be hold on November 10th. It will be presented the status of GENSER and the new build structure based on HepMC. Also foreseen a contribution from EvtGen and future plans. - MC Generators tuning: o The web page for reference MC plots and tuning is now completed and ready for use at CERN: http://mcplots.cern.ch. The work is supported by LPCC. - GENSER: o Several new version of generators installed; some are also new generators: pythia 6.115, charybdis2 1.0.3, tauola++ 1.0.2, hej 2010-09-26, powheg-box 197. For some the build structure was missing and it had to be generated from scratch. There're now roughly 30 different generators nowadays installed in GENSER. o The build structure has been revisited to provide a parallel installation tree of the various generators, making use of the latest version of HepMC-2.06.03 (recently announced and including some minor bug fixes compared to the previous version). o Currently looking at porting all generators to MacOSX (64 bits), and the porting is almost complete. --------------------------------------------------------------------- ==== Fluka - Alfredo - No report. --------------------------------------------------------------------- ==== Geant4 - John - Releases: o Release 9.4 on December 17th. Some of the developments expected for this release are reported below. The release cycle is scheduled to start on November 5th. - Developments expected in 9.4: o Tuning and improvement of FTF hadronic model using HARP-CDP and exclusive channel data; it has been corrected observed cases of excessive production of pions, in particular at energies below 10 GeV. This improves the transition with the Bertini model, making the FTFP_BERT physics-list even more competitive. o Further improvement to the pre-compound and evaporation (CIEMAT) hadronic models. o Further improvements to the Bertini cascade (from SLAC and FNAL teams), reducing memory consumption and now providing a prototype interface to the improved Geant4 pre-compound and evaporation models. o Currently working on the development of a precise safety distance estimation in geometry navigation; also reviewed and improved the overall 'check_mode' verbosity for debugging purposes. - On going developments: o In response to a request from ALICE, a first version of cross-sections for anti-ions (d, t, He) has been calculated and the code to treat them is being implemented. Will then move to the realization of the physics models. o The migration to a new design for 'component' interfaces of hadronic cross-sections is being prepared. - Infrastructure: o The process for migrating the Geant4 source repository from CVS to SVN has been prepared, and it is now planned for January 2011. The tools required to ease the migration, to collect tags and to enable testing are under preparation and almost complete. o The procedure adopted for benchmarking the latest releases and latest monthly development tags has been improved and refreshed. - Validation: o Extended validation with 14.6 GeV Abbott data, HARP-CDP data. o First comparisons have been made with UrQMD ion-ion generator of Binary ion cascade and FTF models (visitor contribution). o Re-iterated systematic comparison of energy fractions into particles produced by the different hadronic models on recent public and development releases. - Feedback from experiments: o ATLAS will evaluate the 9.4 release with a view to use it for production next year. Growing interest in simulating exotic particles, including monopoles. o CMS reports producing 750 million events using Geant4 9.3. o ALICE is testing recent monthly development releases. - Meetings: o The 15th Geant4 Workshop was held at ESTEC on October 4-8th. Many the presentations illustrating the recent development also planned for the December release. One notable item was the agreement to redesign hadronic cross-sections using a component interface. Also, the draft outcome of the architecture review undertaken in the past months was presented to the Collaboration; a set of priority items for revision are being prepared. o Several contributions to the MC2010 conference in Tokyo on October 17-20th: Overview of Geant4 Hadronic Physics; Development of the Fritiof Model; New native QMD model; INCL Intra-nuclear Cascade and ABLA De-excitation Models; Electromagnetic Physics Developments for Ion Therapy Applications; Developments In Pre-Equilibrium and De- Excitation Models; Modeling Radiation Chemistry and Biology; Geant4 Electromagnetic Physics Models and Interfaces; Validation of Geant4 Hadronic Generators versus Thin Target Data. o Two presentations also made at the CHEP10 conference in Taipei on October 18-22nd, on EM physics and hadronic validation/development. o A Technical Forum dedicated to LHC experiments is being organized for November 16th, to collect requirements and agree on the versions to be supported for LHC production use during 2011. o There will be an INFN tutorial in Italian at LNGS at Gran Sasso the week of November 8th, restricted to INFN employees. --------------------------------------------------------------------- ==== Garfield - Rob (offline report) - Garfield: o Considerable progress in the study on the role of photons in avalanche processes and the onset of discharges. A paper will likely be ready for publication submission by the end of the year. - Meetings: o RD51 collaboration meeting in two weeks in Bari. o The program for the "RD51 Simulation School" (January 19th-21st, 2011) has been largely finalized. Contribution to the school likely to focus on MDTs. --------------------------------------------------------------------- ==== AOB Next meeting scheduled for January 11th.