Simulation Project Leaders Meeting 14-May-2008 ==== Attendees John Apostolakis (JA), Alfredo Ferrari (AF), Alberto Ribon (AR), Rob Veenhof (RV), Gabriele Cosmo (GC). Excused: Pere Mato (PM). --------------------------------------------------------------------- ==== Minutes - The summer student assigned to physics validation will start in July for two months. A fellow has been requested for working on Flugg and related test-beam analysis. --------------------------------------------------------------------- ==== Simulation physics validation & framework - Alberto - Meetings: o In the meeting of April 2nd, there has been a presentation by V.Uzhinskiy on the validation of Monte Carlo models for hadron-hadron, hadron-nucleus and nucleus-nucleus interactions related to the Fritiof model, where a common problem seems to be in the treatment of di-quark fragmentation; it was reported about work in progress for improving the Fritiof implementation in Geant4; the combination Fritiof-Binary_Cascade (FTF_BIC physics list) where the usage of the parameterised model is considerably reduced, seems quite interesting and promising, experiments have started to study it. Other presentations from V.Grichine and V.Ivantchenko on biasing of optical photons in PbWO4 crystals and status of standard EM in Geant4 respectively. o In the meeting taken on May 7th, there has been a presentation by T.Carli on an energy scan analysis in the ATLAS Barrel Combined test-beam, scanning from 1 GeV up to 230 GeV, to analyse the energy response and resolution as function of beam energy; the QGSP_BERT and FTF_BIC physics lists were used; FTF_BIC showing a smoother distribution; discontinuities for mean pion response are observed at 5 GeV (where there is a transition between Binary and Fritiof) and 12 GeV (suspected to be related to the beam line geometry); QGSP_BERT has discontinuities at 9 GeV and 25 GeV where the different models interconnect. A suggestion coming from this analysis is to try to improve the overall matching of the models at the transition regions. Another presentation was by G.Folger on improvements in quasi-elastic interactions in Geant4, with fixes now available in the last Geant4 patch 9.1.p02; most coming from feedback by NA61 (problem with high energy secondary in p (29GeV) + C, resulting in an interference between elastic and quasi-elastic) and from HARP-CDP (problem with low energy secondary in 8.9 GeV proton/pion on Beryllium, resulting in an unphysical bump in the angular distribution, corrected by adding Fermi motion in quasi-elastic); a presentation of the HARP-CDP group at the next physics validation meeting (June 18th) on this subject is planned, would be interesting to know the entity of the effect induced by the corrections applied to the data. Another presentation was by S.Kunori on baryon interactions in CMS, where abnormal high Pt particles are produced in Geant4 from 2TeV lambdas at low angles from generators; the problem has been identified in the Geant4 high-energy parametrised model. S.Piperov reported about updates in the CMS 2006 H2 test beam simulation; energy resolution in the simulation results too optimistic with respect to the data; by including layer by layer variations in the scintillator seems to improve a bit; also trying to compare their Birks quenching parameterisation with the one used in L3 analysis, and the results agree; addition of Coulomb barrier in the Bertini model shows little effect; adding Bremsstrhalung of high energy pions (currently under development in Geant4) improves matching with data. - Extension of ATLAS HEC analysis with Fluka: o Have been trying using Flugg for larger production and met some problems; a bug was identified and fixed in G4Polycone in Geant4; another problem shows up and appears related to Flugg and more investigation is required. If new manpower will be assigned (a fellow has been requested), this will become part of the work plan. --------------------------------------------------------------------- ==== Generator services - Alberto - Meetings: o A meeting on MCDB was taken on March 18th with positive feedback. - MCDB: o Small tests for exercising automatic upload of event files in MCDB have succeeded and MCDB is now ready to be used by CMS; a first integration in the CMS software has already been done. Need now to perform a large production testing. - GENSER: o In contact with the author of MC-Tester, a tool currently used for comparing Tauola decays; it is planned to extend its use also for comparing decays (tau and B decays) in different versions of the generators. o New versions of the generators have been included. - HepMC: o A proposal for a new release (2.04) has been made; it mainly consists in the introduction of units treatment in HepMC. Two approaches are under discussion, having 'UNKNOWN' as default for units or force to clearly specify the units; the latter supported by the Durham group. Experiments expressed the wish to have the possibility of keeping their default unit choice. o The Beta release has been announced already and useful feedback was received mainly from the Durham group; currently under discussion for integration in the final release. The plan is to have the release out as a stable release by end of May or beginning of June at the latest. It was not discussed the status code of particles for this release; following the HepEvt rule for now. --------------------------------------------------------------------- ==== Fluka - Alfredo - Releases: o A major release is planned for July, where the new neutron library will be included (library which will be also sent to the Nuclear Agency of Energy). A 5-10% CPU penalty in an high energy shower, depending on the different configurations, is expected. The new release will also include a new neutrino-nucleus generator. - Currently working on damage prediction (DPAs) into the code, feature which will not be yet part of the new release; the idea to help predicting how long a hardware piece can survives to radiation; of great interest for LHC components. - Still working with CNGS; it was finally realised the origin of the hidden misalignements in the beam line detected after inversion of the polarity; the problem was essentially due to the earth magnetic field, and has now been corrected in the simulation. - Meetings: o Course at CERN for the last week of June: 30% of extra people who asked for participation; now making a waiting list. Maximum attendance is for 23 people (limited by the available terminal for the hands-on sessions); a video link with few more people at CERN (no fee requested) will eventually be set up; the ratio physics/software is 40/60. o Another course planned for end September in Paris. --------------------------------------------------------------------- ==== Geant4 - John - Feedback from experiments: o Fixes needed by CMS, ATLAS are included in patch-02 for release 9.1, which was just released. A candidate version for the patch was provided beforehand for evaluation. o Atlas reported that in one channel they were experiencing 1% event failure due to stuck tracks in the geometry since adoption of 8.3.p02 (released in February 2008); under investigation. - Releases: o A new patch to 9.1, 9.1.p02, has been released and including some fixes to address feedback from LHC experiments; a fix for resolving sharp peaks in spectra reported by the HARP-CDP group consisted of the addition of Fermi motion in quasi-elastic in QGS; also included a correction in the neutron module for materials whose isotope composition provided by the user where not treated correctly by HP processes; other fixes affect elastic scattering, where a software bug resulting in sharing a table of cross sections by elastic and quasi-elastic was corrected (feedback from NA61); a request from CMS for having very short-live particles tracked in the detector is also included in the patch. - Developments: o A problem reported with unphysical particles in high angles identified by CMS in the high-energy parameterised (Gheisha-derived) model, is under investigation. The problem was from the interaction of 2 TeV Lambdas. The area of the code has been found and a fix to the a problem has been made available to CMS for testing. o Improvement of quasi-elastic interactions in Bertini, requested also by CMS, is under development. - Publications and conferences: o A number of papers have been submitted to the IEEE/NSS conference in Dresden, in topics spanning geometry, EM physics and hadronic physics. --------------------------------------------------------------------- ==== Garfield - Rob - Simulation of gas-based detectors is becoming part of RD51. RD51 is in the process of becoming a collaboration, with nearly 50 Institutes, combining GEM, Micromegas, Ingrid, radiation hard detectors, associated electronics, RPC. - Several new people are expected to become involved in the simulation work, in particular an Austrian PhD student, a Marie Curie fellow, people from the MPGD group and perhaps Canada-funded students. - Activities over the last months focussed on electron tracking at the molecular level, taking all known Magboltz cross sections into account. This is needed as a result of the fine granularity of the electrodes in several recent gas-based detectors. A by-product of this is the possibility to simulate detectors that rely on photon production. - Future activities include: o interface with boundary element methods, provided by Indian institutes; o interface with Root, to take advantage of its statistical methods; o interface with Geant4, for which it is expected the help from the Geant4 group. --------------------------------------------------------------------- ==== AOB Next meeting scheduled for June 19th.