New Triggerlist

First offline looks at the data were shown at the meeting:

• Daniel Whiteson showed some plots of L3 global track parameters that looked reasonable (except for the known dip in phi that has also been seen for the last couple of weeks in the vertex examine and offline reco. This is currently being investigated). The cft unpacker is producing a large number of error messages. These are an indication of a serious problem - an incorrect cable map and/or corrupted data. We should cross-check whether or not the vertex examine or offline reco are producing the same errors. Daniel will use a picked Z->ee data sample (from Volker Buescher) and a Z->mumu sample (from Gavin Hesketh) to get an estimate of the L3 global tracking efficiency for high p_t isolated tracks in data.
• Ulla Blumenschein showed event selection statistics based on 350 mark and passed events for CEM(1,10). For the calorimeter filters the rejections look similar to those obtained in the April calorimeter special runs and somewhat lower than those obtained in the L3 tracking special run. This is consistent with the statement that the L1CAL energy scale has been put back to the value it had in April. (When the L3 tracking special run was taken in June the L1CAL scale was about 10% too high.) The overall rejection factor of the .or. of all 5 single electron filters hanging off of CEM(1,10) is around 16. Ulla will repeat these measurements on a larger statistics sample.
• Results on muons were not available due to a new problem of the names of some of the l3fanalyze rootuple variables for the muon unpacker being too long.
  Action items: 1) Jon Hays will make the code change to remove unnecessary characters from the variable names. 2) Jon will follow up to make sure that l3fanalyze is run on the reco farms as part of standard production.

Discussion Items

• The urgent need for cookbook-level documentation for general users of how to run the L3 tracking on data and Monte Carlo was re-stated.
  Action item: Daniel will provide this.
  Daniel pointed out, and we all agreed, that it is a nuisance that users have to worry about this. We urgently need a cleaner solution that takes care of getting the correct smt/cft maps, etc, automatically.
• Profiling, timing studies.
  Han Do is at Fermilab for the next four months to work on L3 infrastructure. He will start of by making some measurements of tool timing that we shall use to target our efforts to speed up the L3 code.
• Josh Dyer is implementing release verification histograms for trigsim, including using l3fanalyze for L3. We need to make sure he is talking to Elliot Cheu about our online examine to avoid unnecessary duplication of effort (such as having to talk to Volker).

Scribe: Terry Wyatt.