 |
 |
To continue: 1) The readout simulation also simulates detector electronics, trigger and DAQ. What comes out of the readout simulation is equivalent to the data read out from the real detector. 2a) You should not need a nonzero run number in MC unless you want to run MC with the run conditions from a specific run; run 0 is supposed to give you "ideal" run conditions (we define some conditions for "run -1" because this ensures those conditions are loaded as a default). The conditions system underlying all this is getting changed, but it will remain true that the default run number is appropriate for generic MC. 2b) HPS-Proposal2014-v7 is the current "official" detector - tested and set in stone with no changes allowed. v8 includes more of the ECal support structure but hasn't been tested, and could still change. For any detector version, use "2pt2" to get the field strength and trigger settings appropriate for 2.2 GeV beam. 3) The readout simulation expects that the events in a file correspond to consecutive 2 ns beam bunches. Pulse pileup and DAQ clocks are simulated based on this 2 ns period. So if you were to run the readout sim on a file where every event contains an A', the readout simulation would simulate pileup equivalent to a 500 MHz rate of A's. To avoid pileup between consecutive A' events, you need to insert enough delay (in the form of beam background or empty events) between the A' events. 4) Norman answered this. 5) The stdhep files represent particles coming out of the target. So we simulate interactions in the target (which is centered at the origin), and write out the particles that hit a scoring plane at some positive Z; this Z is different in different generators. 6) The layers used, etc. are set in the tracking strategy. The default is this: http://java.freehep.org/svn/repos/hps/show/java/trunk/tracking/src/main/resources/org/hps/recon/tracking/strategies/HPS-Full.xml and you could make a different strategy file, compile it into hps-java, and point the reconstruction to it using the strategyResource tag in the steering file, like here: http://java.freehep.org/svn/repos/hps/show/java/trunk/steering-files/src/main/resources/org/hps/steering/recon/HPS2014OfflineTruthRecon.lcsim On Fri, 19 Sep 2014, Graf, Norman A. wrote: > Hello Mikhail, > Let me just start with your first question. > > slic/Geant4 does a very good job of propagating particles through material and > fields and calculating energy depositions along the way. > The readout simulations emulate all of the steps in the detector which occur after > an energy deposition happens in the sensitive detector material. For the silicon > tracker this includes, but is not limited to: > > drift and diffusion of charge carriers in the silicon > charge collection on readout strips > electronic readout response modeling > electronic readout noise modeling > overlaying contributions from past and future particles > > In order to retain flexibility in any of these processes we do not code that into > slic. Instead, for every G4Step in a sensitive tracking detector we write out the details of > that step in a SimTrackerHit > http://lcio.desy.de/v01-11/doc/doxygen_api/html/classEVENT_1_1SimTrackerHit.html > That information is used during the "digitization" step of the reconstruction. > > > For the ECal we do not care about the details of all of the secondary particles in > the shower, so we aggragate the contributions of all particle steps in each of the > crystals. That output is written to a SimCalorimeterHit > http://lcio.desy.de/v01-11/doc/doxygen_api/html/classEVENT_1_1SimCalorimeterHit.html > > (note that for dedicated studies one can actually save much more detailed information, > but that is not done by default. > > Norman > > -----Original Message----- > From: [log in to unmask] [mailto:[log in to unmask]] On Behalf Of [log in to unmask] > Sent: Friday, September 19, 2014 9:20 AM > To: hps-software > Subject: MC reconstruction procedure > > Dear Colleagues, > > I am trying to understand the reconstruction process, but I need some information from the experts. > > 1) First of all, a naive question: why do we need the "readout simulations"? I guess SLIC is capable to do this job much better than we ever can develop. May I ask what is saved in the SLIC output then? > Energy deposition in various detector pieces only? > > 2) Do we need a non-zero Run number in MC data, and if so where to set it? Is the following (a small part of) output of recon program acceptable, are all parameters defined for run 0 or even -1 (?): > "No input files in XML file. > Got ConditionsEvent with run: 0 > Reading calibrations calibSVT/base for run: 0 Use this calibration from run -1: calibSVT/default.base Reading calibrations calibSVT/tp for run: 0 Use this calibration from run -1: calibSVT/default.tp Loading the SVT bad channels for run 0 File daqmap/svt0.badchannels was not found! Continuing with only QA bad channels Loading fieldmap for run 0" > > > 2) Which is the "officially definitive" detector geometry table for upcoming experiment? "HPS-Proposal2014-v8"? > > 3) what exactly does pile-up simulation: > org.hps.users.meeg.FilterMCBunches? What do you mean "insert empty bunches"? If they are empty why inserting them? What kind of pile-up is intended here: of particle hits with detector (SVT) noise? of particles from different bunches? > > 4) if I run reconstruction directly from SLIC output using "/org/hps/steering/readout/HPSTrackingDefaults.lcsim" method, I get something that is missing completely "EcalClusters" (although ECalHits" > are present); there are some "MatchedTracks", but with zero momentum components while D0, phi, omega, TanLambda and Chi2 are not zero. > By the way I still missing the description of D0, phi, omega, TanLambda? > Could you point out where these are described? > > 5) why the StdHep files linked at confluence page are generated at vertex (0,0,0.1)? > > 6) How can I change the Track reconstruction parameters: in particular the minimal number of Layers required (are there other parameters)? > > Best Regards, > Mikhail. > > > > ######################################################################## > Use REPLY-ALL to reply to list > > To unsubscribe from the HPS-SOFTWARE list, click the following link: > https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1 > > ######################################################################## > Use REPLY-ALL to reply to list > > To unsubscribe from the HPS-SOFTWARE list, click the following link: > https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1 > ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the HPS-SOFTWARE list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1