Talks

• Status report on L3 tracking studies: Daniel Whiteson.
  Daniel showed updated studies of the rejection of standalone track filters applied to single electron and single muon triggers (using 100K events in the special run taken for L3 tracking) He proposed adding track quality cuts to improve the rejection for the single track filters. The proposed cuts would be to require at least 10 axial hits, which essentially means at least two smt hits, and to require that tracks have stereo information. These requirements reduce the number of fake tracks by about a factor of two.
  After the meeting we agreed on the following thresholds for standalone track filters to be put into the trigger list:

  L1 trigger	p_t cut (GeV)	expected rejection
  CEM(1,10), CEM(2,5)	one track > 25	150
  CEM(1,10), CEM(2,5) followed by L3 filter E_T > 12 GeV + shower shape	one track > 12	300?
  mu1ptxwtzxx_fz	one track > 10	45
  mu1ptxwtzxx_fz	two tracks > 3	16

  Terry will sit down with Elizabeth on Monday and get these implemented in version 8.00 of the trigger list. The axial hit requirement will be added with the current code version. Daniel will implement a change to the track filter for p12 to allow the stereo requirement to be made.
  Daniel also showed some studies of hit and track errors. Pull distributions show that the errors on individual hits look reasonable, but that the track parameter errors look to be overestimated by a factor of 5 or so.
• Shower shape cuts for electrons: Ulla Blumenschein.
  Ulla compared the efficiency of the shower shape cuts used in level 3 with those made offline. The study uses Z->ee data. The conclusion is that the L3 cuts are 100% efficient for high p_t isolated electrons selected by the current offline cuts. It would be nice to repeat these studies using Z->ee events selected with less stringent calorimeter shape and isolation cuts, but with matched tracks to keep the background down. The new, improved shower shape cuts will not be ready for p12.
• Upgrades to the l3fanalyze infrastructure: Jon Hays.
  Jon has made a lot of improvements to the infrastructure for l3fanalyze. In brief:
  • The list of tools to be analyzed can now be configured with a text file.
  • The text file can be automatically generated from the trigger list.
  • d0tools will be modified to make use of this facility automatically.
  • This facility will be used to get l3fanalyze running on the reco farm.
  • The DebugInfo chunk now has a word that distinguishes "online" and "offline" results.
  • It is now possible to run a job to compare "online" and "offline" results in the same job.
  • Robert has modified the unpackers for the tracking detectors so that the DebugInfo output is very much smaller than before.

Discussion Items

• Status of p12 preparations
• Thumbnails and the urgent need for L3 documentation:
  We outlined some of the issues to be discussed seriously next week. PLEASE think about this before next Wednesday:
  • What should be the content of the L3 thumbnail?
  • Should the L3 physics results be re-created when the thumbnail is unpacked?
  • How do we go about getting documentation available for the L3 physics results and roottuple contents?
  • How do we go about getting detailed writeups of the algorithms employed in L3?
  • How do we go about getting a simple example analysis job, with a user analysis package that loops over all of the physics results and makes a few simple histograms?

Scribe: Terry Wyatt.