Simulation Project Leaders Meeting 18-March-2005 ==== Attendees John Apostolakis (JA), Alfredo Ferrari (AF), Witek Pokorski (WP), Alberto Ribon (AR), Rob Veenhof (RV), John Harvey (JH), Pere Mato (PM), Gabriele Cosmo (GC). Excused: Paolo Bartalini --------------------------------------------------------------------- ==== Minutes The meeting was mainly dedicated to the preparation of the AA internal review which will take place from March 30th to April 1st. A skeleton of the agenda for the simulation project was prepared. Presentations will have to be included in the official agenda by each subproject leader well before the review. Slides of the workplan (milestones for 2005 and proposed longer term activities) should be provided to GC as soon as possible. GC prepared a page for additional documentation to be provided to the reviewers. --------------------------------------------------------------------- ==== Simulation physics validation - Alberto - AR presented the main themes of activity he would like to propose at the review for the long term. Among these are: o Identification of more simple benchmarks to test the MonteCarlo engines; o Provide help to experiments to use the implemented infrastructure for validation for other test-beam studies; o Collect data a-la JetWeb useful for tuning and validation of detector simulations; RV mentioned that a similar project is going on for nuclear ion-beam studies. AF noted that it would nice to have a centralised place on the web where to put links to data. o Invest some time in trying to transform the existing LHC experiments setups to standalone setups to be used for regular MonteCarlo validation at every new release of the software; It was noted the difficulty in trying to achieve this, mainly due to the amount of code involved which will require regular maintenance and the fact that most setups may be very specific to each experiment. o In the more longer term, define and put in place a systematic procedure for tuning and validation of simulation including event-generators. The idea has to be discussed with people in the experiments and in the generator services project. AF noted that it would be something very useful although it is not always possible to avoid to have specific tunings in some cases. GC pointed out that a systematic Validation will certainly be required for the new C++ generator packages. - A note describing the usage of FLUGG for test-beam studies has been prepared by AR and WP and is now under revision. - The next Physics Validation meeting will be on Wednesday, March 23rd where presentations on a study on calibration constants for simulation in ATLAS calorimeter test-beam are expected. Another meeting is scheduled for May 4th in which a presentation on radiation studies in LHCb is expected. --------------------------------------------------------------------- ==== Generator services - Paolo Attached is an extract from the memo that PB sent before the meeting: - Manpower: The list of the MCDB Developers & GENSER Integrators for Q1+Q2 2005 is reported below. GENSER User support from MC experts cannot be guaranteed in the period March 12th-March 28th. The same will happen in the period of May 1st-May 10th. User support is currently representing around 50% of the GENSER integrator duties. The lack of integrators in the periods mentioned above can represent a serious problem. For this reason, although distribution area and tarballs are ready, the first production quality relase of GENSER is postponed of two additional weeks and scheduled on March 28th. - GENSER: The release of GENSER_1_0_0 is scheduled on March 28th (MILESTONE). With respect to the report given to the previous SL meeting, some additional time was required to wait for bug fixes from authors (Herwig 6.507) and to redefine the way to assign tags to the sub-packages. Alexandre Sherstnev will join on March 28th. Priority will be given to work on the connection between MCDB and the first root of the validation framework (collection of use cases of GENSER). - MCDB: No activity during the last 5 weeks. Sergey Belov just arrived. He will concentrate on the improvement of the web interface. He will work in conjunction with A.Sherstnev for the integration between MCDB and the validation framework. - The next LCG Generator meeting will be dedicated to the proposal for the development of a NRQCD model for prompt J/Psi production to be implemented in Pythia 6.3. The contact person for this activity is V.Vagnoni (INFN Bologna & LHCb). The plans for the relase of GENSER_1_0_0 were presented to the LCG Generator meeting of February 24th. The release will be presented to one of the first AA meeting in April (when there will be one time slot available). ++ MCDB Developers & GENSER Integrators for Q1+Q2 2005. - S.Makarychev ITEP (Moscow) from 2004 to 31/01/2005 Genser (30%) + MCDB (70%) - I. Kachaev IHEP (Protvino) from 11/01/2005 to 11/03/2005 Genser - S. Slabospitsky IHEP (Protvino) from 14/02/2005 to 28/02/2005 Genser - S. Belov JINR (Dubna) from 06/03/2005 to 09/05/2005 MCDB - A. Sherstnev SINP MSU (Moscow) from 28/03/2005 to 28/04/2005 Genser - L. Dudko SINP MSU (Moscow) from 10/05/2005 to 10/06/2005 MCDB - M. Kirsanov INR RAS (Troitsk) from 10/05/2005 to 31/07/2005 Genser --------------------------------------------------------------------- ==== Fluka - Alfredo - Development and testing for the new release is going on and since one week the new Fluka is running without PEMF. Currently under testing. A slight penalty in initialisation time is observed. Working on trying to reduce it. - A lot of activity was going on recently on exotic processes for ions, including a model for pi-pair production for ions. - A detailed plan from now to the release is in place. - A test library of the new Fluka-Beta will be provided to WP to test. - Alice is very close to certify their geometry and validate it against the native Fluka geometry. --------------------------------------------------------------------- ==== Simulation framework - Witek - GDML: a new release (2.1.0) will be announced soon. There's strong request for it from several customers. A project for a technical student to work on GDML will be defined. - FLUGG: a note for the usage of FLUGG for validation has been prepared. A generic example showing how to run FLUGG and produce hits a-la Geant4 has been implemented and could be included in the FLUGG distribution, with the idea to use it also for the new testbeam studies. It is planned for the future to evaluate the usage of VMC. - Python: the project web page (http://lcgapp.cern.ch/project/simu/framework) has been reorganised; it will include soon also descriptions on how to use Python/PyReflex to steer a Geant4 application. --------------------------------------------------------------------- ==== Geant4 - John - A TSB meeting of the Geant4 Collaboration was held on March 17th. It was in good part dedicated to define the release plan for 2005. A minor release of Geant4 has been scheduled for the end of June; one of the developments it will include is the new implementation for the particle definitions. The list of the planned developments for 2005 will be posted soon from the web; they will include planned developments for requirements collected in the past months and technical-fora. - A Technical-Forum was held on March 8th: open requirements were reviewed and new one collected, among these an issue on reading 4-momentum of pre-decay particles. Other oustanding problems were reported, among which the need for "silent" running for production use. - Experiments are moving to use release 7.0. The patch released in February includes fixes important for the usage in the LHC experiments. - Work is going on for radiation studies (LHCb), to find the best suitable hadronic physics-list. --------------------------------------------------------------------- ==== Garfield - Rob - From 4-14/2, RV has been in Breckenridge for NA60. Quite some feedback from the presentation made, in particular regarding the interference pattern between rho and omega which would explain the excess signal claimed by E325. - Forthcoming visit to Napoli and a lecture series in early May in Cagliari. Also received an invitation for a talk at a conference in Bari on detection of keV energy photons in gas detectors. - Other SFT activities at CERN: 1. In collaboration with COMSOL, suppliers of FEMLAB, shape functions for their 2nd order Lagrange elements have been added along with the inverse derivatives & Jacobians of the isoparametric coordinates that are associated with these elements. Both 2D and 3D field maps from FEMLAB can now be processed, including 2D field maps with rotational symmetry such as those encountered in TPC field cages. These are now being used by several groups, in Florida and at Panda. Given the quality of the field maps that are computed by FEMLAB, as well as the price of the Ansoft products, it seems natural that users at CERN will move to this program. 2. Similar functionality has been added for hexahedral and degenerated hexahedral (pentahedral) elements. These are used by the VF programs such as Tosca. Inverting the isoparametric coordinates for these elements is difficult (if at all possible) analytically, and is therefore done in iterative fashion from a nearly isoparametric seed. 3. The documentation for Garfield has been linked to Savannah and a bug reporting system has been added. Gradually starting to get more and more feedback from users. 4. A paper written in collaboration with Rachel Avramidou (ATLAS) regarding the properties of the gas used in the ATLAS MDTs has come back from the referees and we're almost done with the clarifications and corrections requested by the referees. 5. Refereed a paper on integral equations for IEEE. This is of particular interest since this technique is a priori far preferable to the finite element technique more often used in HEP. Spent some time with the authors on understanding the advantages. The paper has now been approved for publication. Expected to continue the collaboration with the authors of this paper (in India). 6. Fixing the diffusion in non-Gaussian transport areas is advancing more or less as scheduled and should be finished in about a month. This point concerns in particular detectors like GEMs. --------------------------------------------------------------------- ==== AOB Next meeting will be in one month.