sPHENIX tracking discussion

["Anderson, Derek [PHYSA]" <dmawxc AT IASTATE.EDU>]

Hi Xin,

Apologies for the delayed response! But thanks for the feedback! I've in-lined some responses below:

 -- Derek

> Shall we evaluate these tracking qualities (chi2, pT resolution) vs. nHits as well?

Definitely! That'll be a very good check!

> And I may have missed your study before, do you have the momentum resolution vs. pT performance plot (maybe not only mean, but also the mean value)?

So far I've only put together 2D plots of pTreco vs. pTmc (e.g. the color plots buried under the normal/weird track markers in the right panels of figure 6), but I can put together a plot that more explicitly shows the momentum resolution vs. pT.

> Slide 24, what is the unit on your X-axis? mm? And are the silicon hits required in tracks for producing these figures?

The x-axis here is in micrometers, and that's correct: these were made requiring silicon hits (they were from before the iterative tracker was implemented back in December).

> Slide 29, Is Delta_DCA the error of DCA? I know often people plot DCA / DCA_error and defined as chi2primary (used in KFParticle).

Ah! I didn't know that! But that's correct: delta-DCA is the uncertainty on the DCA.

> Slide 31/32/33, these distributions are peaked at very small values. I think it is because you have often finite values for pT, eta, phi. Is it better to plot   pT_err / (pT_rec - pT_truth) or (pT_rec - pT_truth) / pT_err.

Ah! I see! I'll make sure and present them that way on subsequent iterations.

> Slide 37/38, similarly, I am curious whether the silicon hits are used in tracking.

Yep! These require silicon hits (also note that the x-axis is wrong here).

> Can we access the reconstructed hit information on track to tell whether it is the real MC hit coming from the associated MC track?

This is definitely a question for the experts: I believe so, but I'm not completely sure...

> Slide 54/55, I am wondering what is the difference compared to slide 37/38? Are these using tracking with silicon hits?

The only difference here is that the x-axis scale is the completely wrong (these are from back when I first starting to look into things). But these are indeed tracks with silicon hits.

 -- Derek

---

From: Xin Dong <xdong AT lbl.gov>
Sent: Wednesday, February 1, 2023 7:33 PM
To: Anderson, Derek [PHYSA] <dmawxc AT IASTATE.EDU>
Cc: sphenix-tracking-l AT lists.bnl.gov <sphenix-tracking-l AT lists.bnl.gov>; Anthony Frawley <afrawley AT fsu.edu>
Subject: Re: [Sphenix-tracking-l] Tracking software meeting Wednesday February 1 at 10 am ET

 

Dear Derek,

Thank you for your very comprehensive slides. I am sorry that I couldn't attend the meeting for discussion. I have a few questions about your slides and would appreciate your clarifications.

1) Slide 7, you showed nicely that most of the weird tracks lie at sector boundaries. Presumably these tracks will have less number of hits on track compared to others, yielding worse resolution. Shall we evaluate these tracking qualities (chi2, pT resolution) vs. nHits as well? If the tail in the pT distribution is just due to the worse resolution that yields due to few nHits, I would view this is not really an issue we should worry (but just feature?).  And I may have missed your study before, do you have the momentum resolution vs. pT performance plot (maybe not only mean, but also the mean value)?

2) Slide 24, what is the unit on your X-axis? mm? And are the silicon hits required in tracks for producing these figures?

3) Slide 29, Is Delta_DCA the error of DCA? I know often people plot DCA / DCA_error and defined as chi2primary (used in KFParticle).

4) Slide 31/32/33, these distributions are peaked at very small values. I think it is because you have often finite values for pT, eta, phi. Is it better to plot   pT_err / (pT_rec - pT_truth) or (pT_rec - pT_truth) / pT_err. In this way, if the tracking has a good estimation of the error, the distributions will be more like chi2 distributions too.

5) Slide 37/38, similarly, I am curious whether the silicon hits are used in tracking. Can we access the reconstructed hit information on track to tell whether it is the real MC hit coming from the associated MC track? This information can be helpful to determine the ghost hit rate, particularly on silicon MVTX detector.

6) Slide 54/55, I am wondering what is the difference compared to slide 37/38? Are these using tracking with silicon hits? I think then one would need different scales for the DCA axis to look into its distributions.

Thank you and Best Regards

/xin

On Wed, Feb 1, 2023 at 10:47 AM Anderson, Derek [PHYSA] via sPHENIX-tracking-l <sphenix-tracking-l AT lists.bnl.gov> wrote:

Hi all,

Just following up: please find my slide from today attached!

 -- Derek

---

From: sPHENIX-tracking-l <sphenix-tracking-l-bounces AT lists.bnl.gov> on behalf of Anthony Frawley via sPHENIX-tracking-l <sphenix-tracking-l AT lists.bnl.gov>
Sent: Tuesday, January 31, 2023 2:25 PM
To: sphenix-tracking-l AT lists.bnl.gov <sphenix-tracking-l AT lists.bnl.gov>
Subject: [Sphenix-tracking-l] Tracking software meeting Wednesday February 1 at 10 am ET

 

Hi All,

We will have our usual tracking software meeting tomorrow, Wednesday February 1, at 10 am ET.

The zoom link is:

 https://fsu.zoom.us/j/98126117840

The indico page is:

https://indico.bnl.gov/event/18283/

Cheers

Tony

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

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================