• 3 Dec 2007
    • TODO
      1. To get B->Jpsi_digit.root in CMSSW_1_6_6
        1. Trial in Lxplus : Failed. I can't give a credit about generated root files within Lxplus. -> Done.
        2. Trial in MIT TIER : Merged gen.cfg, sim.cfg and digireco.cfg and get genreco.cfg. and got gendigi.root. Tried to analyze with ana.C(root macro) -> Error
          ssh jkim@cgate.mit.edu
          ssh -Y hibat0001 // For interactive access
          scramv1 project CMSSW CMSSW_1_5_4
          cd CMS_1_5_4/src
          eval `scramv1 runtime -csh`
          cd Bevents/GEN_DIGI
          cmsRun BBar2Jpsi.cfg // output file : "/net/pstore01/d00/scratch/jkim/bgenreco1000.root" 
      2. L1 test
        • With Olga's(Dongho's) code :
          cvs co -r V00-00-02 UserCode/OlgaKodolova
        • Check hltBToJPsiToMumu.cfg,
      3. Mixing
        1. Send bgen_reco.cfg to Yetkin to make HI events mixing with HYDJET background sample.
        2. He will match the vertex. Normaly I used "Configuration/StandardSequences/data/VtxSmearedGauss.cff" but maybe this time "IOMC/EventVertexGenerators/data/VtxSmearedFlat.cfi" is needed.
          include "IOMC/EventVertexGenerators/data/VtxSmearedFlat.cfi"
             replace VtxSmeared.MinX = 0.0001
             replace VtxSmeared.MaxX = 0.0001
             replace VtxSmeared.MinY = 0.0002
             replace VtxSmeared.MaxY = 0.0002
             replace VtxSmeared.MinZ = 2.
             replace VtxSmeared.MaxZ = 2. 
          Above things are also in the HYDJET.cfg.( HeavyIonMCProduction )

  • 4 Dec 2007
    • TODO
      1. Read Mixing module twiki page and send email to Yetkin.(reco.cfg & vertex smearing)
      2. Modify analysis_l1l2.cfg and cmsRun with diversity source file(digi, raw, reco, etc.) to get trigger efficiency. -> digi file is OK. Dongho's analyze_l1l2.cfg is working well. -> Charge, Eta, Pt, Phi...
        untracked InputTag GMTInputTag = l1GmtEmulDigis // For Level-1 trigger
        InputTag  CandTag = hltL2Muons // For Level-2 trigger 

  • 8 Dec 2007
    • Post Analysis Note on iCMS

  • 10 Dec 2007
    • Got GRID certificate from CERN
    • VO Registration is needed
    • Produced BBarToJpsi?_DIGI 3000 events in MIT -> /tmp/jkim -> /castor/cern.ch/user/j/jkim/SecondaryJpsi : If I want use /tmp/jkim, I must login the same machine. (ex. ssh lxplus237) Checked hltBToJpsi path. It works. Need to think about preselection.

  • 11 Dec 2007
    • TODO
      1. How to deal with "preselection" -> Check with 500 events.
        • Does mufilter play a role for preselection? Yes or No?
        • Is another filter needed after mufilter for preselection? Yes or No?
        • Is anothet cfg file needed only for preselection? Yes or No?
      2. Look into TestMuL1L2? module
      3. Look into Mixing Module
      4. Read OL1 documents

  • 13 Dec 2007
    • TODO
      1. Check Dong Ho's question to L1 tirgger group
        • lowqualityFlag information in the L1Trigger?/CSCTrackFinder/data/CSCTrackFinder.cfi (Dong Ho)
          • 0 = none, 1 = DT/CSC overlap, 2 = ME 1/1a, 3 = both, 4 = everywhere
        • The eta coverage is also limited for single muon triggers in pp collisions because ME1/1a (the inner radius part of the ME1/1 chamber) does not participate in the trigger, and hence we do not get a good sagitta measurement. Pt resolution suffers, and rate would be high. But the CSC Track-Finder anyway sends low quality candidates from regions where there is only coverage by ME4, ME3, and ME2. So no reason to change the flag you list from its default of "4" in the CSCTrackFinder?.cfi. What would have to change is how GMT or GT uses such low quality muons. By default only the dimuon trigger allows them at high eta. So this is a question to GT/GMT experts. ( Darin )
        • in fact all CSCTF candidates become GMT candidates (if they are not suppressed by an equivalent DTTF one) and all are passed to GT. The GMT candidates carry the quality information with them. The way low quality bits are set in the GMT candidates is described in the note CMS IN 2004/006 page 4. Candidates with quality lower than 4 (see the note) get suppressed in the standard single muon GT conditions. So the way to extend the single muon trigger to a wider range would be to add a single muon GT algorithm with no quality restriction. Probably the most practical way is to simply look at GMT candidates(from L1MuGMTReadoutCollection?) in unfiltered data. ( Ivan )
      2. Check CMSSW/HLTrigger/Configuration/data/common/HLTSetupCase1.cff for L1 setup.
        • include "Configuration/StandardSequences/data/L1Emulator.cff" -> include "L1Trigger/Configuration/data/L1Emulator.cff"
        • include "Configuration/StandardSequences/data/L1Extra.cff" -> include "L1Trigger/Configuration/data/L1Extra.cff"

    • Q1 : Can filters be in "OR" mode?(Answered in Hyper News : Event Filter Development)
      • If they are independent HLT triggers, then this is automatically ORed: an event is accepeted if any of the HLT triggers accepts it (== OR). Also, you could write ONE filter whose filter condition is a Boolean OR of the conditions you want - this maybe appropriate for very tightly connected "single" triggers, using the same quantities to trigger on.(Martin Gruenewald)
    • Q2 : parameters, MinPt?, MinEta?, etc. are working well? -> Ask to Evt Filter group.
    • Q3 : CMSSW/L1Trigger/Configuration/data/L1Emulator.cff in CMSSW_1_6_6
      include "L1Trigger/Configuration/data/L1MuonEmulator.cff"
      include "L1Trigger/Configuration/data/L1CaloEmulator.cff" 
      include "L1Trigger/GlobalTrigger/data/l1GtEmulDigis.cfi"
      sequence L1Emulator = { L1CaloEmulator, L1MuonEmulator }
      // GT emulator will be included by default in 170 series
      // For the L1 decision in 160, run L1Extra sequence
      // defined in L1Trigger/Configuration/data/L1Extra.cff
    • Q4 : CMSSW/L1Trigger/Configuration/data/L1Extra.cff in CMSSW_1_6_6
      include "L1Trigger/L1ExtraFromDigis/data/l1extraParticles.cff"
      include "L1Trigger/L1ExtraFromDigis/data/l1extraParticleMap.cfi"
      sequence L1Extra = { l1extraParticles, l1extraParticleMap }

L1Trigger?/L1ExtraFromDigis/data/l1extraParticles.cff -> L1Trigger?/L1ExtraFromDigis/data/l1extraParticles.cfi -> module l1extraParticles = L1ExtraParticlesProd
Looking into L1Trigger?/L1ExtraFromDigis/src/L1ExtraParticlesProd.cc

  • 15 Dec 2007
    • TODO
      • Look into L1Trigger?/Configuration/data/L1Emulator.cff -> L1Trigger?/Configuration/data/L1MuonEmulator.cff
        # CSC Trigger
        include "Geometry/CSCGeometry/data/cscGeometry.cfi"
        include "L1Trigger/CSCCommonTrigger/data/CSCCommonTrigger.cfi"
        include "L1Trigger/CSCTriggerPrimitives/data/l1CscTpgEmulDigis.cfi"
        include "L1Trigger/CSCTrackFinder/data/l1CscTfTrackEmulDigis.cfi"
        include "L1Trigger/CSCTrackFinder/data/l1CscTfEmulDigis.cfi" 
      • LowQualityFlag? is in L1Trigger?/CSCTrackFinder/data/l1CscTfTrackEmulDigis.cfi

  • 17 Dec 2007
    • TODO
      1. Check the result plots(eta) when LowQualityFlag? in 'L1Trigger/CSCTrackFinder/data/l1CscTfTrackEmulDigis.cfi' is changed from 4 to 1. -> Changed or not? This works for testing LowQualityFlag? value 4 is suitable for HI collisions. If I change the value to 1, eta range is changed to |eta|<2.1?

      • LowQualityFlag? = 1

      • LowQualityFlag? = 4
      • Result : LowQualityFlag? is not working. Both eta plots are exactly identical. How can I fix this? frown -> Ask it to Ivan.

  • 20 Dec 2007
    • TODO
      1. Ivan answerd like this :
        1. Dimuon L1 acceptance extends up to eta<2.4 but the pt threshold must be not higher then 3GeV.
        2. For single muons, the proper way to extend the L1 acceptance to eta<2.4 is to add to the trigger menu a single muon algorithm without quality restriction and with pt threshold not higher than 3GeV.
        3. you do not need to manipulate these quality bits. If what you need is a single muon trigger extending to eta<2.4, you just need to add an algorithm to the L1 trigger menu in a way I described. I think, there should be anyway a special trigger menu for HI collisions. These things should be discussed in the Trigger Studies Group.
      2. Analyze GenParticleCand? with Dong Ho's JPsiCandAlgoMoon?.cfg -> This is a revised file in PhysicsTools?/CandAlgos/test/ by Dong Ho. I used MCSingleParticleFilter? in generation, but it didn't cut |eta|<2.4. I'll change it to MCParticlePairFilter?.
      • J/Psi's eta plot using MCSingleParticleFilter? : Preselection didn't seem to be applied.
        root -l jpsi_CandAlgo.root
        Events.Draw("JPsisCand.data_ .p4_.eta()");
        Events.Draw("JPsisCand.data_ .p4_.pt()");
        Events.Draw("JPsisCand.data_ .p4_.mass()");
        Events.Draw("muons.data_ .p4_.eta()");
        Events.Draw("muons.data_ .p4_.pt()");
        Events.Draw("muons.data_ .p4_.mass()");

-- JiHyunKim - 20 Dec 2007

Topic attachments
I Attachment History Action Size Date Who Comment
PDFpdf CMSDec07.pdf r1 manage 203.4 K 2008-01-22 - 18:56 UnknownUser  
GIFgif JPsi_eta_500_SigFilter.gif r1 manage 11.2 K 2007-12-20 - 16:55 UnknownUser  
PDFpdf SecondaryJpsi_7Dec07.pdf r1 manage 2575.6 K 2008-01-22 - 18:57 UnknownUser  
GIFgif eta_l1_3000_qu1.gif r1 manage 11.7 K 2007-12-20 - 16:30 UnknownUser  
GIFgif eta_l1_3000_qu4.gif r1 manage 11.7 K 2007-12-20 - 16:31 UnknownUser  
Edit | Attach | Watch | Print version | History: r5 < r4 < r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r5 - 2008-02-14 - JiHyunKim
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding KoreaCmsWiki? Send feedback