News

• Certification for p13 online.
  Certification for p13 online was delayed by a number of problems. The final one was that a rebuild was required to fix problems with L3MEt. The change to p13 online in L3 and trigger list v10.0 must be made simultaneously. The run coordinator and trigger board chair have decided these changes should not be made with so many critical personnel away for the holidays. We shall procede with the final stages of p13 certification, including tests with trigsim of the v10 trigger list, although the earliest that this is likely to run online looks like early in the new year.

  The latest roottuples using p13.05, global_CMT-9.42, useL2=yes, are at: /work/lappland-clued0/yann/p1305/rootuples/d0trigsim_Global9.42_168522_1/L1L2L3global.root and toolstats.root.

  The latest roottuple using p13.06, global_CMT-10.0, useL1=yes, is at:
  /prj_root/826/alg_24/dhan/rootfiles/p130600_global_10.0_useL1yes_168522_1/L1L2L3global.root and toolstats.root.
  Please take a look!

Talk

• Progress towards certification of L3 MuoLocal and MuoCentralMatch: Martin Wegner.
  Considerable progress has been made towards certification of L3 MuoLocal and MuoCentralMatch. The following issues/open questions remain:
  • Relative to offline medium muons with matched central tracks, the efficiency of L3 MuoCentralMatch is around 80% in p13, whereas it was around 95% in p12.
    • We need more events for plots from special run. 20K events processed whereas many more (80K?) available.
    • The p12->p13 change appears to be due to lower relative efficiency for L3 MuoLocal wrt offline MuoLocal.
    • Is this because p13 offline is more efficient or is it finding more background?
  • J/Psi -> mumu sample with both muons matched to central tracks offline has now been processed successfully through RECO and trigsim:
    • Efficiency to find at least one L3 MuoCentralMatch is only 80%! There is a clear discrepancy with 80% per muon efficiency found in the single muon study. Need relative number of events with 0, 1, 2 local muons, 0, 1, 2 central tracks and 0, 1, 2 MuoCentralMatches.
  • We really need a sample of Z -> mumu events to measure the performance on a very clean sample of high pt muons. A standalone L3 LocalMuon (i.e., without requiring a central track match) would be immensely valuable to provide some redundancy wrt L3 central track filters, even with a 50% efficiency per muon.
  • Exactly which geometry is used by L3Muon in p13. Does it have the 3cm shift?
  • Things that it would be nice to do, but not necessarily before L3Muon is regarded as available to run online:
    • Study of tuning the .rcp parameters for L3Muon to try to recover some of the "lost" efficiency wrt offline.
    • L3 MuoCentralMatch doesn't know about relative polarity changes between solenoid and the toroids. Does this affect the matching efficiency?
    • The L3 MuoCentralMatch turn on curves for pt>5 GeV are very slow. This appears to be due to L3 central track resolution being much poorer than offline. Needs someone (preferably other than Martin) to investigate!
    • It would be great if someone (preferably other than Martin) could use the J/Psi -> mumu sample sample to study L3 central track efficiency and resolution.

Discussion Item

• Yann Coadou (tau), Martin Wegner (muon) and Per Jonsson (b-tag) had taken a quick look at the new package l3ftrack_phys_results. The general consensus was that it shouldn't be too much work to implement the necessary changes and that we should aim to get these modifications made in January for p14.
  Action item for all physics tool authors: by mid-January make new package for each tool to hold the physics_results.
• Action item for all members of the L3 filtering group: have a Merry Christmas and a Happy New Year!
• Next L3 meeting is scheduled for Wednesday 8th January 2003.

Scribe: Terry Wyatt.