RUN 2B LEVEL-1 TRIGGER MEETING
			      14-Feb-02

News: Darien Wood
-----
o Upcoming Reviews
  - week of Mar 25 - Temple
  - mid Apr - mid Jul - Lehman
    . Note add after the meeting: probable date is now 25-27 June

o Technical Design Report
  - update: design, cost, schedule, simulation
  - cost est's: more detail, cost justification
  - ready ~3 weeks before review

o L1Cal Technical Milestones
  - 21 Feb: agreement on ADF-TAB interface
  - 25 Feb: pieces of TDR
  - 27 Feb: 1st draft TDR
  - 4  Mar: Doc's ready
  - end Mar: eng. meeting
  - May: L1Cal workshop

o MRI
  - submitted Jan 24
  - $2.6M = $1.3M(equip) + $1.3M(personnel)
    . no contingency
  - Start date: 1 Sep 2002


Run 2b Level-1 Simulation Issues: Mike Hildreth
--------------------------------
o Overriding Goal
  - quantify benefits of trigger upgrade (Higgs sensitivity)

o Trigger List
  - have put together a strawman L1 "Higgs" trigger list that can
    currently be simulated
    . these will be Run 2a triggers
  - Z --> b-bbar missing
  - Will calculate rates based on this
    . apply reduction factors for Run 2b
  - Will try to get sliding windows into Trigsim
    . already available standalone
    . this requires making special Ntuples that gives accurate TT
      energies (rounding problem)
  - L1CTT will not be ready
    . but manchester group has this going offline
  - Cal-Trk matching will also not be possible
  - Should we add triggers (like 2 track tau-trigger) that we'd like
    to use in 2b that aren't in the 2a list???
  - Discussion:
    . need to emphasize the 2b trigger rate: is the upgrade good
      enough?

o Combined Offline Analysis
  - use Run 2b silicon (Lisa & Flera) + Run 2a trigger
    . allows (in principle) to write down combined b-tagging eff, for
      example 
  - note: this is still in its early stages

o All this should be ready in a week or two


L1CTT Simulation: Terry Wyatt
----------------
o Simple simulation package written to allow studies of various
  schemes.
  - not included in tsim_l1cft, but takes the same input
  - ready to be released into CVS

o Tests of Infrastructure
  - standard doublet algo implemented w/ standard eqn's --> identical
    results to tsim_l1cft
  - Yuri's singlet-doublet algorithm implemented --> some
    discrepencies wrt CDR exist
    . need to sort these out w/ Yuri

o Manchester plans for Study
  - start with a scheme treating all 16 layers as individual hits
    (with no vetoes from other layers)
    . try various combinations 16/16, 15/16, etc.
    . will get equations from Graham
  - order these eqn's in integral acceptance
    . study effects of cutting out low-accept eqn's
  - currently doing this with the standard doublet eqn's
    . noise trigger rate scales ~ linearly w/ number of eqn's
    . looking into how eff. changes
    . with 2-3% reduction in eff. can reduce no. eqn's by 50%
    . results not particularly sensitive to N(min-bias)
    . impact param tracks: no big difference from 0-impact param trks
      > studied this with taus
      > not yet w/ b's, but have done prelim work on checking the
        effects of shifting beam spot
  - may need to use different sets of equations for different types of
    triggers 
    . could keep no. of eqn's used as a sort of quality criterion

o FPGA Algorithms
  - need to start to pin down how complicated the algorithms will be
  - can only do this when singlet eqn's are available

o Detailed talk will be given in the next meeting
  - transparencies already available - will post on web