sPHENIX discussion of software

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

Also available on agenda: https://indico.bnl.gov/conferenceDisplay.py?confId=1162

 

Announcement:

We use new time slot and room for future sPHENIX simulation meeting:

Tue 1-3 PM at 510-2-160 Conference Room by default

 

John L. - HCal simulation thoughts

 

Summary: nice overview on possible path forward on HCal simulation

 

Page4:

Ed: digitalized form important to extract Time-of-Flight

Jin:

1.       We have a simple HCal tower class, and John's plan give great overview of functionality we want to include

2.       Some correction need to be done in the

 

Page 8:

We agree that we should move forward on adopting a policy on organizing GitHub code development with larger group of people. Code reivew and sign-off procedure need to be defined.

 

Action Item [Martin and John L.]: make tutorials on policy management from other GitHub projects, so we can learn and adopts

 

Liang - Simulation summary

 

Summary: for the current considerations on the Hcal simulation analysis

 

Jin- Production considerations

 

Summary: some thinking on organizing formal productions.

 

An existing code doing simple HCal towering: 

1.       Start with macros/g4simulations/G4_HcalOut_ref.C
https://github.com/sPHENIX-Collaboration/macros/blob/master/macros/g4simulations/G4_HcalOut_ref.C

2.       PHG4HcalCellReco group PHG4Hits together to make Cells (collection of G4hits within a set of scintillation tile)

3.       RawTowerBuilder convert PHG4HcalCellReco into tower objects.

4.       RawTowerBuilder save RawTowerContainer onto the DST nodes, which contains a set of active tower in one event

5.       Each tower is described by RawTowerv1 , which currently contains the eta + phi bins and G4 total energy deposition

We probably want to make a new tower object which know its location + G4 total energy deposition + visible energy deposition for the short term HCal studies.

 

Birk's law code implementation for EMCal:

1.       EMCal stepping function calls for visible energy calculated for each Geant4 tracking step:
https://www.phenix.bnl.gov/WWW/offline/doxygen_sPHENIX/html/d8/dcf/PHG4SteppingAction_8cc_source.html#l00088

double light_yield = GetVisibleEnergyDeposition(aStep);

2.       PHG4SteppingAction::GetVisibleEnergyDeposition use the standard Geant4 G4EmSaturation module to convert Geant4 ionization energy loss to visible energy with Birk's law correction based on G4 standard Birk constant list.

3.       Then for SPACAL PHG4SpacalSteppingAction::UserSteppingAction() , three energy is tracked and saved to the output PHG4Hitv6 (then later used for towering):

a.       hit->set_edep(...) save total Geant4 truth energy deposition

b.      hit->set_eion(...) save Geant4 ionization energy deposition

c.       hit->set_light_yield(...) save visible energy which is ionization energy deposition x Birk's law correction

We probably want to implement similar thing in PHG4HcalSteppingAction::UserSteppingAction()

 

 

______________________________

 

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, May 18, 2015 4:13 PM
To: phenix-nextmid-l; 'sphenix-software-l AT lists.bnl.gov'
Cc: sphenix-hcal-l AT lists.bnl.gov
Subject: [New Time] Tue 1-3PM sPHENIX simulation meetings in HCal workfest week

 

Dear All,

 

This is reminder that our weekly sPHENIX simulation meeting is now moved to Tue 1-3PM. The default room at BNL is the 2-160 conference room. We will try this time slot starting tomorrow.

 

For tomorrow, it is a special meeting in joint with the GSU HCal workfest with HCal simulation planning as the focus. The agenda is

https://indico.bnl.gov/conferenceDisplay.py?confId=1162

The modification password is 1008

 

We will use the HCal workfest phone bridge for the meeting:

 

To join the Meeting:

 

https://bluejeans.com/721140482

 

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: 721140482

 

Cheers

 

Jin

 

 

______________________________

 

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: Friday, May 15, 2015 3:28 PM
To: phenix-nextmid-l; sphenix-software-l AT lists.bnl.gov
Subject: [minutes] Friday 10AM sPHENIX simulation meetings

 

Also available on the agenda page:

https://indico.bnl.gov/conferenceDisplay.py?confId=1147

 

Announcement:

·      We summarized the TODO items for EMCal and jet simulations. New volunteers are welcomed for the many TODO items as highlighted below

·      Next sPHENIX meeting on May 19 Tue 1PM, where HCal workfest summary will be presented.

 

Jin - EMCal

 

Summary: give a summary of the current EMCal simulation status, on-going R&D on projection and near term TODO list.

 

Page 7: Sasha confirm that a parameterized EMCal performance was used in ePHENIX LOI plots, rather from G4. Nils plan to study it with G4.

 

Pag14: Current 2-D EMCal simulation is on-hold waiting for the parent engineering layout to finish

 

Action Item [Jin]: Finish implementing 2-D EMCal simulation

 

Page 17:

Nils: from CALICE experience, it is important to study the dependence of Birks's law correction VS G4 range cut

 

Page 18:

Can we make the back end of the EMCal module to be flat in z direction, so the zig-zagging edge is flatten? We should consider that.

 

Action Item [Jin]:

·      provide a prototype of Geant4->ROOT geometry exportation based on TGeoManager

·      Update performance plots with Birk's law

 

Wish Items  [Manpower needed]:

·      Study performance VS photon transportation efficiency and electronics noise

·      Does 2D-tapering introduce too much variation in sampling fraction from front to back side of the module? Wave-hand argument say it should be small effect. But need to quantify.

·      Tower-by-tower shower shape analysis based on modern machine learning algorithms

·      Try anti-kT for EMCal clustering with R = Molière radius. Exploratory whether it works. (Mike: the evaluator provide convenient tool to quantify the performance)

·      Tool chain to quantify the performance with physics deliverables, e.g. Upsilon RAA error bar.

 

 

Mike - Jet

 

Summary: proposing re-organize jet objects and give a plan on the coming jet studies.

 

Page 2:

Also plan to have truth jet on DST

 

Page 3:

A nice plan of TODO list with open item highlighted. Contribution are welcomed.

 

Action Item [Mike]:

·      Jet storage structure today.

·      Improving evaluation process.

·      Jet evaluator

·      SVX reco modulization

 

Wish Item [Manpower needed]:

·      Import over ATLAS jet reco method (Ali or Anne?)

·      SVX reco improvement  (?)

·      SVX+TPC evaluation (Alan?)

·      SVTX trackk object redesign(?)

·      Reco modulization, e.g. improve Kalman filter(?)

 

Page 4-6:

Action Item [Mike]:

·      We plan to proceed with Plan-B and let user try it out. We will collect user experience to make final conclusion in future meetings.

·      To remedy the disadvantage for Plan-B in term of hard coded jet name based finding, we plan to give each map program-searchable flags, like radius, source and algorithm code. So a user code can iterate through all jet-maps, and find the jet satisfying given criteria without knowing the string name for the jet.

 

Page 7:

Plan A is preferred by Mike and Jin

Dennis: question on the natural of the 4x4 covariance uncertainty. Mike: can be computed and may reserve it for possible use in the future.

 

Page 9:

Dennis: useful to include some of the feature (e.g. B-fraction, jet shape) to Jet object itself.

Jin: it could be resident in a stl::map of (property ID, float value)  inside the standard sPHENIX jet objects. So we do not blow up Jet object into 10 versions (as our PHG4Hit).

Mike: It is important to put the first version in, to allow immediate studies and feedbacks. Then we further improve it.

 

Wish Item [Manpower needed]:

·      Particle Flow Jets (Javier)

·      Bottom Jets (Dennis, Peter, Mike using 2nd vertex?)

·      Jet Rejection with shape cut (?)

·      Jet Purity  with Jet Modification (Ali?)

·      Recoil jet background subtractions (?)

·      Raw level tune of fast sim (?)

·      Geant4 unfold of jet spectrum (?)

 

 

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: phenix-nextmid-l-bounces AT lists.bnl.gov [mailto:phenix-nextmid-l-bounces AT lists.bnl.gov] On Behalf Of Huang, Jin
Sent: Thursday, May 14, 2015 4:23 PM
To: phenix-nextmid-l; sphenix-software-l AT lists.bnl.gov
Subject: [reminder] Friday 10AM sPHENIX simulation meetings

 

Dear All,

 

This a reminder for the next sPHENIX simulation meeting, Friday 10AM-12PM. Currently we have two confirmed talks on EMCal and Jet. More topics are welcomed too.

 

Agenda: https://indico.bnl.gov/conferenceDisplay.py?confId=1147 , edit password is 1008

Location: for BNL residence we will meet in conference room 2-219.

Phone bridge: https://bluejeans.com/233203754/browser
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: 233203754

 

Also to remind that the second next meeting will be on Tue May 19, 1PM-3PM in joint with GSU HCal workfest.

 

Cheers,

 

Jin

 

 

______________________________

 

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: Wednesday, May 13, 2015 11:02 AM
To: phenix-nextmid-l; 'sphenix-software-l AT lists.bnl.gov'
Subject: [new time slot] coming sPHENIX simulation meetings

 

Dear All,

 

I am writing to update on the plan for the coming sPHENIX meetings. Please note the change of the time slots for all the meetings:

 

Following the tracking summary/planning in the last meeting, we will have two meetings continue the summary/planning for calorimetry and jet studies:

·       Fri May 15, 10AM – 12PM, Core week sPHENIX simulation meeting.
In this meeting we will focus on the EMCal and Jet studies. The meeting time was shifted one hour later to accommodate the PSB meeting
https://indico.bnl.gov/conferenceDisplay.py?confId=1147

·       Tue May 19, 1PM-3PM, joint meeting with HCal workfest summary session
HCal will be the focus as you may expect
https://indico.bnl.gov/conferenceDisplay.py?confId=1162

 

Through the summer, we will try to move the meeting to a new weekly time slot of Tuesday 1PM-3PM, starting the week of May 19. Feedback are welcomed.

 

In the future meetings, besides continuing developing the physics program, we will also ramp up the effort in performance study for finalizing the detector designs, and in improving the sPHENIX software environment for long-term use. New contributions of topics and suggestions are always welcomed too.

 

Cheers,

 

Jin

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________