sPHENIX tracking discussion

[Anthony Frawley <afrawley AT fsu.edu>]

,Hi All,

I realized that I said something that is wrong at the meeting this morning, when we were discussing the onset of poorer Y(1S) mass resolution from our tracking code. Here is the correction.

The attached slides (made last week) show the Y(1S) mass resolution from our current code, obtained with different tracking options. In all cases, truth vertexing was used. Normal Acts chain seeding consists of PHTpcTracker + TPC tracklet vertex association + Si track seeding + truth vertexing + TPC-Si track matching.

The results are:

1) Normal Acts chain seeding + Acts track fit:                              95.0 +/- 2.6 MeV (used to be ~ 85 MeV)

2) Genfit track propagation + Genfit track fitting:                       73.8 +/- 1.7 MeV

3) Normal Acts chain seeding + Genfit final fit                             81.3 +/- 1.4 MeV

4) Normal Acts chain seeding + Acts fit + Genfit final fit:           82.9 +/- 1.9 MeV

So, using Genfit track propagation with a Genfit final fit DOES produce a significantly better mass resolution than using our normal track seeding with a Genfit final fit. Also, adding a final fit with Genfit after Acts fitting produces an improvement in mass resolution of ~15%.

The difference between 2) and 3) is interesting.

Tony

Attachment: Investigate_mom_resolution_loss.pdf
Description: Investigate_mom_resolution_loss.pdf