sPHENIX discussion of software

[Alan Dion <alan.dion AT gmail.com>]

Hi Jin,

There have been some small changes to the tracking code in the TPC branch which probably do not affect the performance for the silicon tracker, but it would be good to be absolutely sure if we want to merge everything back into one branch.  For this, we should devise some regression tests : a macro which produces performance plots so we know the merge didn't make anything worse.  Exactly what those plots should be can be discussed at the next meeting.

Cheers,

Alan

On Wed, Aug 12, 2015 at 4:31 PM Huang, Jin <jhuang AT bnl.gov> wrote:

Hi Alan

> Merging the TPC branch is definitely planned.  I think I will wait until the 

> evaluator is stable.  In my private branch I am using a hacked up 

> evaluator in the meantime that I don't want to merge in to master.

Great to know the plan. I agree with the need for the the development branch on TPC and special procedure on the evaluation. Meanwhile, we try to keep the isolation time low for the development branches, which we intended to eventually merge with the master. As you know the master branch evolves fast, and it would be harder and harder for a development branch to merge as it gets diverged from the master. 

> I doubt that the EMCAL matching will be very different between the TPC 

> and silicon trackers, so if computing time is a problem we can use 

> results from the silicon simulation for eID performance.

I agree in the leading order they would be the same as the ID-performance is dominated by the calorimeters background and resolution. Nevertheless, at some later time, I would be curious to check the effect of the reduced electron radiative tail on the Emcal with the TPC and thinned SVX options.

Cheers,

Jin

______________________________

Jin HUANG

Brookhaven National Laboratory
Physics Department, Bldg 510 C
Upton, NY 11973-5000

Phone: 631-344-5898
______________________________

-------- Original message --------

From: Alan Dion <alan.dion AT gmail.com>
Date:08/12/2015 12:20 (GMT-05:00)
To: "Huang, Jin" <jhuang AT bnl.gov>, "Soltz, Ron" <soltz1 AT llnl.gov>
Cc: sphenix-software-l AT lists.bnl.gov
Subject: Re: [Sphenix-software-l] [minutes] sPHENIX simulation meeting, Tue Aug 11th 1-3PM @ 2-187

Hi Jin,

Merging the TPC branch is definitely planned.  I think I will wait until the evaluator is stable.  In my private branch I am using a hacked up evaluator in the meantime that I don't want to merge in to master.

I doubt that the EMCAL matching will be very different between the TPC and silicon trackers, so if computing time is a problem we can use results from the silicon simulation for eID performance.

Cheers,

Alan

On Wed, Aug 12, 2015 at 12:08 PM Huang, Jin <jhuang AT bnl.gov> wrote:

Hi Ron

Thank you for the update!

A head up, regarding the next simulation meeting, it would be possible that we use a different time slot or use email update, depending on the schedule for the dry run for the emcal review on Tue. I will keep the list posted on it.

For TPC-only simulation, you are welcome to uodate the TPC git branch with the master and run your studies. If we need to run a joint simulation between TPC and the calorimeters (say for quantify the AuAu eID performance with reduced material before EMCAL), then I think it would he best that we merge TPC branch to master and run the simulation production centrally.

Cheers,

Jin

______________________________

Jin HUANG

Brookhaven National Laboratory
Physics Department, Bldg 510 C
Upton, NY 11973-5000

Phone: 631-344-5898
______________________________

-------- Original message --------
From: "Soltz, Ron" <soltz1 AT llnl.gov>
Date: 08/12/2015 11:49 AM (GMT-05:00)
To: "Huang, Jin" <jhuang AT bnl.gov>
Cc: sphenix-software-l AT lists.bnl.gov
Subject: Re: [Sphenix-software-l] [minutes] sPHENIX simulation meeting, Tue Aug 11th 1-3PM @ 2-187

Hi Jin,

Here’s a quick answer to the first question from Chris: Yes

* We do want to run separate tracking simulations with TPC options

Alan has been working on this, and Jeff, Takao, and I have been using

We can make this a discussion point for the next meeting.

-Ron

On Aug 11, 2015, at 1:41 PM, Huang, Jin <jhuang AT bnl.gov> wrote:

This minutes is also available on today’s agenda: https://indico.bnl.gov/conferenceDisplay.py?confId=1186

 

All: simulation plan for coming reviews

 

Page 3:

Crag: what is the plan for simulating test beam

Chris: we will put test beam detector geometry in Geant4. It will be a separated study comparing to this effort.

 

Many: Do we want to run separate tracking simulation with TPC options?

No TPC expert around for the meeting. Need to track down with Alan and Ron.

 

Action Item [Tony]: plan to run Upsilon simulation to verify the current SVX implementation in GitHub master still maintain good performance (Upsilon mass resolution fo ~100 MeV) after many recent updates.

If so, the current version of SVX in cylinder geometry will be used in the reference simulation production, while tracking group continue the development of laddered SVX.

 

Page 4:

Peter & Chris: there is still some problem on read-in the vertex from HEPMC event.

Action Item [Chris]: check event generator vertex implementation.

 

Page 5:

Tony: Recent test showed an error of duplicated G4 hit ID complained from evaluator

Action Item [Tony & Chris] : check whether the problem still show up after the newest nightly build and local build.

 

Abhisek: one possible way to specify the production setup is to tag the production macros on GitHub

 

Page 6:

Chris: Embedding with two G4Hit input stream not yet implemented

Action Item [Need volunteer]: module and procedure to enable embed one PHG4Hit DST stream (usually Jets or single particle) to the other stream (usually HIJING).

 

Page 7:

Chris: tracking code should be verified and get tagged now

 

Discussion:

Chris: need man power and plots

Action Item [John Haggerty]: plan to summarize the basic detector plots needed for the pre-CDR

Action Item [Manpower Needed]: analyze the simulation data.

 

Chris: should start a test production as soon as possible. We are close to have everything ready (now pending on verifications for the implementation of projective EMCal and SVX tracker).

Action Item [Jin]: verify the new 2DSpacal branch and merge it to master. Then a test production can start

 

 

Abhisek: single particle in Hcal

 

Page 2:

Chris: another ntuple module??

Abhisek, John and Jin: No. 
This module emphasize on a standard way of reading the DST node. Different analysis will require different histograms or ntuples, which people could produce by building their code on top of this example module.

 

Page 3:

Chris: the phi dependence probably comes from the beat between frequency of the steel plate the frequency of phi angle scan point. To avoid such beat, We can either adjust N phi-bin to a large number or an integral of .

Action Item [Abhisek] : please run the material scan again following Chris’ suggestion. It will be great that we verify it and be sure of the new material distribution before the production starts.

 

Page 7:

Craig: energy resolution too good?

Action Item [Abhisek] : 1. Check towered/clustered energy resolution which would take into account of lateral leakage. 2. Cross check with Liang's study for the last generation of G4 HCal setup.

 

Page 8:

John L. and Jin: expected proton shower energy is the kinematic energy of the input proton. They leads to this apparent divergent at lower total energy points. 

Action Item [Abhisek] : please try kinematic energy as the x-axis for proton leakage/resolution study

 

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: sphenix-l-bounces AT lists.bnl.gov [mailto:sphenix-l-bounces AT lists.bnl.gov] On Behalf Of Huang, Jin
Sent: Monday, August 10, 2015 11:53 AM
To: sphenix-l AT lists.bnl.gov; sphenix-software-l AT lists.bnl.gov
Subject: [Sphenix-l] [reminder] sPHENIX simulation meetings, Tue Aug 11th 1-3PM @ 2-187

 

 

Dear All,

 

This is reminder of our weekly sPHENIX simulation meeting Tue Aug 11^th 1-3PM. For residents at BNL, we will meet in the 2-187 conference room (for this week only).

 

We will start the meeting by discussing software readiness and simulation plan for coming reviews and pre-CDR writing. More updates and contributions are welcomed too.

 

The agenda page is at https://indico.bnl.gov/conferenceDisplay.py?confId=1186  , for which the modification password is 1008

 

To join via Browser:

https://bluejeans.com/377457648/browser

 

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: 377457648

 

 

Cheers,

 

Jin

 

 

 

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

_______________________________________________
Sphenix-software-l mailing list
Sphenix-software-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-software-l

_______________________________________________
Sphenix-software-l mailing list
Sphenix-software-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-software-l