Compute RecHit resolution in CMSSW job

This twiki contains the instructions on how to create trend plots of the RecHit resolution as a function of a given variable from CMSSW jobs. Resolutions are computed from the SimHit-RecHit residuals, hence are available only for simulated events.


The first step is the production of ntuples using the EDAnalyzer NewStdRecHitNtuplizer (available here).

An example of including the fragment in a GEN-SIM-RECO sequence can be found here

Warning, important Resolutions vs. η shoud be computed using a pointlike beamspot (especilaly along the z axis) pset fragment.

The output ntuple contains two trees of pixel residuals:

  • PixelNtuple all RecHit
  • Pixel2Ntuple on-track RecHit only

Warning, important Currently some of the variables are still empty

The script to process the ntuples can be found here. An inline help is available:

Usage: [options] 
  -h, --help            show this help message and exit
  -f INPUT_ROOT_FILENAME, --file=INPUT_ROOT_FILENAME input root file
  -o, --on-track        use on track clusters (default is all clusters)
  -g, --gauss           gaussian fit of residuals (default is RMS)
  -e ENTRIES, --entries=ENTRIES number of entries 

The script produces a set of TCanvas (saved as pdf files) with trend plots of several variables as a function of η, cotg(β) and z, and a set of TCanvas saved as ROOT files used as input to the pretty printing script (rmsVsEta_rz.root, rmsVsEta_rphi.root, rmsVsZeta_rz.root, rmsVsZeta_rphi.root, rmsVsCotgBeta_rz.root, rmsVsCotgBeta_rphi.root). All ROOT objects are also saved in PlotResHistos_All_RMS.root

Final display of results (overlaying curves, pretty printing)

The script to produce the final (pretty printing) plots of the RecHit-SimHit resolution as a function of η, cotg(β) and z can be found at here.

Example of the output (PhaseI: 100x150 μm2, threshold of the digitizer 2000 e) Warning, important PLOTS FOR ILLUSTRATION ONLY

foo_rphi_extY_ageing.png foo_rz_extY_ageing.png

Qave is the average charge of the clusters from primaries in the bin abs(η) < 0.2 corrected by the local angle Q⋅cosγ (e.g. scaled to normal incidence angle). For each RecHit the charge of the associated cluster, corrected for the incidence angle Q⋅cosγ, is then compared against Qave.

-- ErnestoMigliore - 02 May 2014

Edit | Attach | Watch | Print version | History: r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r3 - 2014-05-05 - ErnestoMigliore
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback