sPHENIX tracking discussion

[Michael Joseph Peters <mjpeters AT mit.edu>]

Hi Joe,

It's not clear to me right now what's driving that. Currently working on collecting the missed tracks in that region and finding things they have in common. At first glance, though, since ignoring the silicon hit requirement doesn't seem change the shape of the curve very much at low pt (as shown in the lower plot), I'm not expecting a matching issue to be the main cause.

Thanks,

Michael Peters

---

From: Osborn, Joe <osbornjd AT ornl.gov>
Sent: Thursday, June 17, 2021 5:51 PM
To: Michael Joseph Peters; christof.roland AT cern.ch; Anthony Frawley
Cc: sphenix-tracking-l AT lists.bnl.gov
Subject: Re: [Sphenix-tracking-l] CA Seeding with Propagator Update

 

Hi Michael,

 

That looks really good! Just out of curiosity, do you understand why the “turn on” around 1 GeV is much smoother after including the propagator? E.g. it seems like the propagator actually hurts the efficiency between 0.5-1 GeV or so. Or is it a matching issue?

 

---------------------------

 

Joe Osborn, Ph.D.

Postdoctoral Research Associate

Oak Ridge National Laboratory

osbornjd AT ornl.gov

(859)-433-8738

 

 

From: sPHENIX-tracking-l <sphenix-tracking-l-bounces AT lists.bnl.gov> on behalf of Michael Joseph Peters via sPHENIX-tracking-l <sphenix-tracking-l AT lists.bnl.gov>
Date: Thursday, June 17, 2021 at 5:42 PM
To: christof.roland AT cern.ch <Christof.Roland AT cern.ch>, Anthony Frawley <afrawley AT fsu.edu>
Cc: sphenix-tracking-l AT lists.bnl.gov <sphenix-tracking-l AT lists.bnl.gov>
Subject: [EXTERNAL] Re: [Sphenix-tracking-l] CA Seeding with Propagator Update

Hi all,

 

PR (#1218) for this update is now submitted. In the meantime, as requested, here are the efficiencies with ntpc>24, rather than nhits>20:

Since the propagated seeds are on average quite long compared to the original CA seeds, a higher cut on track length increases the improvement in efficiency when the propagator is added.

 

In addition, here's an efficiency plot with only ntpc>24, ignoring any quality cuts or silicon matching cuts:

Thanks,

 

Michael Peters

---

From: Christof Roland <christof.roland AT cern.ch>
Sent: Wednesday, June 16, 2021 3:16 AM
To: Anthony Frawley
Cc: sphenix-tracking-l AT lists.bnl.gov; Michael Joseph Peters
Subject: Re: [Sphenix-tracking-l] CA Seeding with Propagator Update

 

Hi Michael, 

 

thats great progress, cool.

 

just one detail we should fix in the performance plots fpr TPC only. 

We should use a quality cut of at least number of layers divided by 2, so 

long tracks that are split in two show up only once and the shorter part is cut. 

 

cheers

 

    Christof 

 

On 16. Jun 2021, at 02:22, Anthony Frawley via sPHENIX-tracking-l <sphenix-tracking-l AT lists.bnl.gov> wrote:

 

Hello Michael,

 

Nice! The improvement in the number of clusters is particularly impressive.

 

I would not focus too much on the number of track seeds at this stage. PHTpcTracker also  produces a lot more seeds than there are actual tracks. We will have to clean them up after the fact. The top priority now is track finding efficiency, with the highest possible number of associated clusters.

 

Good news!

Tony

 

 

---

From: sPHENIX-tracking-l <sphenix-tracking-l-bounces AT lists.bnl.gov> on behalf of Michael Joseph Peters via sPHENIX-tracking-l <sphenix-tracking-l AT lists.bnl.gov>
Sent: Tuesday, June 15, 2021 8:12 PM
To: sphenix-tracking-l AT lists.bnl.gov <sphenix-tracking-l AT lists.bnl.gov>
Subject: [Sphenix-tracking-l] CA Seeding with Propagator Update

 

Hi all,

 

Yesterday I showed that the propagator was finally beginning to do more good than harm for some tracks, but at the time it still wasn't a uniform improvement. Since then, I believe I've finally found a propagator design that works; tuning is in progress, but here are the results so far:

 

<newprop_compare.PNG>

The efficiency is now a little bit under 90% above 2 GeV (whereas with CA seeding alone it was around 80%). As for nhits:

 

<newprop_nhits_compare.PNG>

There are a lot more tracks produced overall (the tuning is currently pretty inclusive right now, which exacerbates this), and the secondary peak at intermediate nhits is now gone. I'll continue working on optimizing the propagator, but it seems I've finally hit on a workable design.

 

Thanks,

 

Michael Peters

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