sPHENIX discussion of software

["Huang, Jin" <jhuang AT bnl.gov>]

Also available on agenda: https://indico.bnl.gov/conferenceDisplay.py?confId=1924

 

Software update

 

·       Option of GenFit refit of tracks: https://github.com/sPHENIX-Collaboration/coresoftware/pull/193

·       Truth tracking: https://github.com/sPHENIX-Collaboration/coresoftware/pull/192

 

Action Item [Jin/Mike]: review and merge

 

 

Tracking macro - Tony

 

Tony

·       Setup 3MAPS + 4Strip + TPC  (60 layers), only 80% eff though expect higher. Still debugging. 

·       Allow setup variable layers of silicon tracker, therefore support MPS + TPC configuration in the same macro too.

·       Tried copy over Carlos' work where better efficiency is observed. Contacting Carlos to cross check.

 

Sourav/Tony: observed higher efficiency with tracking QA module than plotting from Evaluator. Some difference from RMS VS gauss fit in resolution, translated to eff via 4-sigma momentum matching requirement in efficiency definition. 

 

Action Item [Tony]: complete check and commit macro very soon.

 

Tony: HIJING in tracking condor jobs fails sometime with job types of CPU_Experiment  = “experiment”

Chris: “phenix” jobs support 4GB in mem usage, while “star” slots has much lower limit, which could evicted simulation. Suggest use  
            Requirements    = (CPU_Experiment == "phenix")

If hit limit, use the more limited high-mem slots:  https://wiki.bnl.gov/sPHENIX/index.php/Condor

 

Dennis: what is the speed of HIJING embedding simulation in TPC-based trackers

Tony: could be 1 event per hour if full evaluation feature is enabled

Jin/Tony: much faster (1 event / min ) if only evaluate tracks, if cluster/hit level of evaluation is not needed. Example:

 

	SvtxEvaluator* eval = new SvtxEvaluator("SVTXEVALUATOR", outputfile.c_str());
	eval->do_cluster_eval(false);
	eval->do_g4hit_eval(false);
	eval->do_hit_eval(false);
	eval->do_gpoint_eval(false);
	eval->scan_for_embedded(true);
	eval->Verbosity(verbosity);
	se->registerSubsystem( eval );

 

 

Jet structure TG: FF in G4 - Dennis Perrapalisa

 

Summary: test tool chain with output of FF performance for tracking review

 

Tony: would high pT jet yield be quantified?

Dennis: yes.

 

Jin: what was used in pythia q^hat?

Dennis: q^hat = 45 GeV/c for jets of pT>50 GeV/c

 

Dennis: what should be good definition of good track in cluster counting?

Tony: number of truth cluster matched and Chi2 cut will be very useful in defining efficiency

 

Tony: suggest use a standard resolution curve to define signalized cut of  momentum reconstruction efficiency

Action Item [Tony]: produce standard resolution plot for both MAPS+TPC and MAPS+ Intermediate Tracker + TPC

 

Jin: what’s the cause of low-eff at intermediate pT

Tony: hit - 30 in Hough Module is useful in relaxing cluster requirement for TPC

           PHG4HoughTransformTPC* hough = new PHG4HoughTransformTPC(Max_si_layer,Max_si_layer-6);

Tony: the new standard macro will carry this change.

 

Target plots:

·       Purity and resolution for tracks in jet up to 50 GeV/c

·       Track/jet yield projection at high pT

 

 

HF-jet TG: B-jet tagging via Geant4-track counting - Haiwang Yu

 

Summary: showed performance plot for b-tagging with tracking counting in full Geant4 tracking simulation.

 

Slide3: simulation setup

Jin: pTHatMin = 30 GeV is way too high. Should reduce to 20 GeV or lower for 25 GeV jet.

Action Item [Haiwang]: will update to ptHatMin = 10 GeV while analyze jet of pT>20 GeV

 

Slide 8: purity VS efficiency for MAPS + TPC

Jin: Why purity VS efficiency has an inversed trend in second significant DCA tagging at the low efficiency end?

Haiwang: long lived tail was observed in the light quark jet

Jin: suggest cut those light jets away with a max DCA distance cut

 

Jin: how large is the statistical error in these curves?

Haiwang: 1M event was generated, leave ~1k statistics at the working point after rejection cuts

 

Slide 9: With MAPS + TPC, better performance observed

Tony: expected from better performance of MAPS.

Tony: should not show MIE performance plots for clarity, just for sanity checks

 

Dennis: fast sim is at 20 GeV. G4 sim at 30 GeV. Suggest check consistent region.

Action Item [Haiwang]: use 20 GeV region in pT.

 

Target plots:

·       Update purity VS efficiency plots with comments and new tracking configuration macro. Solidify result for p+p

·       Stretch goal: HIJING embedded performance

 

 

Upsilon TG: embedded e-ID performance - Sasha 

 

Summary: study electron ID in G4 simulated shower embedded into full HIJING events

 

Slide 5: Inner HCal and EMCal cuts are highly correlated

Jin: last generation showed inner HCal was only helpful in anti-proton rejection.

 

Slide 6: rejection VS pT

Jin: slight dependency of for eta-dependence was also observed for rejection VS momentum. However larger eta dependence is expected for rejection VS pT

Tony/Marzia: pT was chosen for the x-axis for ease of correlating with background yield. Nevertheless, plan is to produce eta-pT 2D rejection curves.  

 

Plan:

·       Calculate rejection for anti-protons and kaons

·       Check eta dependence

·       Generate the new background spectrum after applying the momentum and eta-dependent hadron rejection

 

Limiting factor: available high-memory condor slot.

 

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: Huang, Jin
Sent: Monday, August 22, 2016 2:40 PM
To: 'sphenix-software-l AT lists.bnl.gov' <sphenix-software-l AT lists.bnl.gov>; 'sphenix-l AT lists.bnl.gov' <sphenix-l AT lists.bnl.gov>
Subject: [reminder]sPHENIX simulation and software meeting, Tue Aug 23rd, 1PM EST @ 2-219

 

Hello everyone,

 

This is a reminder for a regular sPHENIX simulation and software meeting, Tue Aug 23rd, 1PM EST @ 2-219.

 

We will focus on the preparation of simulation results for the coming tracking review. As mentioned in the last general meeting, many topical groups planned to give simulation updates in this meeting.

 

The agenda page is at https://indico.bnl.gov/conferenceDisplay.py?confId=1924 , for which the common modification password is "1008"

•            For residents at BNL, we will meet at the 2-219 conference room this time

•            To join the Meeting via phone bridge:  https://bluejeans.com/619960505 

•            To join via Phone:

1) Dial:

        +1.408.740.7256

        +1.888.240.2560(US Toll Free)

        +1.408.317.9253(Alternate number)

        (see all numbers - http://bluejeans.com/numbers)

2) Enter Conference ID: 619960505

 

Cheers,

 

Jin

 

 

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________