sPHENIX discussion of software

[Anthony Frawley <afrawley AT fsu.edu>]

Hi Hugo,

I should qualify that I am using Acts tracking, not Genfit.

Tony

---

From: sPHENIX-software-l <sphenix-software-l-bounces AT lists.bnl.gov> on behalf of Anthony Frawley <afrawley AT fsu.edu>
Sent: Friday, January 8, 2021 1:49 PM
To: Huang, Jin <jhuang AT bnl.gov>; Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>; Osborn, Joe <osbornjd AT ornl.gov>; sphenix-software-l AT lists.bnl.gov <sphenix-software-l AT lists.bnl.gov>
Subject: Re: [Sphenix-software-l] Deterioration in upsilon invariant mass resolution

 

Hi Hugo,

Yes, I also am having trouble reproducing this. The pion momentum resolution looks normal. I did check that this was so after the electron drift speedup changes.

Tony

---

From: Huang, Jin <jhuang AT bnl.gov>
Sent: Friday, January 8, 2021 1:44 PM
To: Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>; Anthony Frawley <afrawley AT fsu.edu>; Osborn, Joe <osbornjd AT ornl.gov>; sphenix-software-l AT lists.bnl.gov <sphenix-software-l AT lists.bnl.gov>
Subject: RE: [Sphenix-software-l] Deterioration in upsilon invariant mass resolution

 

Hi Hugo

 

You are one step ahead the revival of the tracking QA, which are supposed to capture stuff like this.

 

Did a quick check on Upsilon mass with the current draft tracking QA module: on the contrary, the mass appears mostly consistent with previous (although there are some shift as reference was GenFit tracking):

https://nbviewer.jupyter.org/github/sPHENIX-Collaboration/QA-gallery/blob/jenkins-sPHENIX-test-tracking-low-occupancy-qa-647-test-tracking_Event100_Sum10/QA-tracking.ipynb#Upsilon-reconstruction

 

This is low occupancy setup, 1 Upsilon embedded into 40 pions,:

https://github.com/sPHENIX-Collaboration/macros/compare/master...QA-tracking-low-occupancy

and built on the last commit on the Clang-Format branch:

https://github.com/sPHENIX-Collaboration/coresoftware/pull/1042

which was rebased to carry PR1039 optimization of PHG4TPCElectronDrit, including 784ae25 :

https://github.com/sPHENIX-Collaboration/coresoftware/network

 

Cheers

 

Jin

 

 

______________________________

 

Jin HUANG

 

Physicist, Ph.D.

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>
Sent: Friday, January 8, 2021 1:11 PM
To: Anthony Frawley <afrawley AT fsu.edu>; Huang, Jin <jhuang AT bnl.gov>; Osborn, Joe <osbornjd AT ornl.gov>; sphenix-software-l AT lists.bnl.gov
Subject: Re: [Sphenix-software-l] Deterioration in upsilon invariant mass resolution

 

Hi Tony,

 

it is low occupancy yes. only one upsilon (so two electrons) per event.

Macro attached. (that you need to modify if you want to run due to own evaluators)

 

Hugo

 

On 1/8/21 11:07 AM, Anthony Frawley wrote:

Hello Hugo,

 

What particles are thrown in these events? They are low occupancy?

 

Thanks

Tony

---

From: sPHENIX-software-l <sphenix-software-l-bounces AT lists.bnl.gov> on behalf of Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>
Sent: Friday, January 8, 2021 12:13 PM
To: Anthony Frawley <afrawley AT fsu.edu>; Huang, Jin <jhuang AT bnl.gov>; Osborn, Joe <osbornjd AT ornl.gov>; sphenix-software-l AT lists.bnl.gov <sphenix-software-l AT lists.bnl.gov>
Subject: [Sphenix-software-l] Deterioration in upsilon invariant mass resolution

 

Hi Tony, others,

with latest code I see a terrible upsilon invariant mass resolution. (while using GenFit toolchain, which has not change for month). This was not the case on december 21st.

Looking back at the comits since then I think I have identified the culprit as being commit 784ae256185aa9c77d4604c59f0cf1dd1cdc3548 (the optimization of PHG4TPCElectronDrit) Plots attached represent the Upsilon inv mass resolution before (InvMassUpsilon_upsilon_truth_micromegas_nominal-old.pdf) and after (InvMassUpsilon_upsilon_truth_micromegas_nominal-new.pdf) the said bunch of commits. 

Similar deterioration is seen for the track momentum resolution and for the cluster residual width (cluster - truth) in the TPC (which is what pointed me to this commit in the first place). Can you double-check ?

Don't hesitate to ask if you need more information from my side.

Best,

Hugo