Agenda 10-Aug-2006 Telephone Meeting

• Notes from 18-Jul-2006 meeting

• Reports from the CERN run (hopefully).
  Power failure/storm damage over weekend 29-July, in particular to failure of PS injection septum meant that most of scheduled ~2 weeks beam time from 31-Jul was lost. Injection attempted p.m. of 7-Aug, data taken on 8-9 Aug, up to 40 GeV e^-, and 60 GeV pions. Beam size is large, ~8x8cm. Data may be difficult to obtain in next few days.

• Reports on data analysis from the May'06 DESY running
  
  • Preliminary estimate of energy resolution - David Hadley
    Reduced background following David Ward approach of previous meeting using several variables in likelihood, then fitted single and double electon peaks to determine energy resolution of ~17%/sqrt(E). Slightly larger than expected resolution attributed to Small leakage, no correction for missing layers event-by-event and no weighting for differing absorber thicknesses.

• Digisim and noise studies (.ppt, .pdf) - Anne-Marie Magnan

• Physics analysis /Pflow studies
  • MAPS clustering - Bradley Hopkinson
    Plans to start working on clustering specifically for MAPS, based on position/density of clusters alone
  • Status of particle flow with PandoraPFA - Mark Thomson
    Summarised current performance and plans to distribute code in near future. Noted change in approach for case where a problematic cluster energy/track momentum are not in agreement: will now attempt to split up the cluster by reapplying whole clustering until acceptable agreement with track momentum obtained. If not possible, the track itself will be dropped leaving cluster alone.
    Significant efforts at optimisation (but not much room for improvement found), reclustering most important at higher energies. Current version of code available at: http://www.hep.phy.cam.ac.uk/~thomson/pandoraPFA, performance as documented in LCWS06 writeup, and today's slides.
    Suggestion was that now code is available several of us should try to use it and evaluate performance in different applications and with e.g. variation of hadronic interaction modelling, etc.
  • ee->vvWW and ee->vvZZ (.pdf, .ps) - Wenbiao Yan
    Summary of ongoing WW/ZZ analysis work, comparing di-jet mass resolutions of W and Z at each of parton level, hadron level and detector level (using PandoraPFA). Found parton shower increases fitted BW width by x2.5 relative to natural W/Z width, while hadronisation/undetected particles are ~10% effects. Have obtained results for WW/ZZ separation through entire analysis chain now.

• MC Production (RHUL, Bham, Manch. 1, Manch. 1) - Fab/Dave B/Nige
  Had met at UCL recently, agreed various details e.g. naming conventions for MC production files for grid storage elements, versions of G4/Mokka and dependant packages for purposes of testing MC generation at each of the three institutes. Testing with same simple test job on each farm using Fab's TBdesy0506 driver. Dave had made some changes to simplify setting of random number seeds with Geant, code distributed and being tested.