sPHENIX discussion of physics

[Anthony Frawley <afrawley AT fsu.edu>]

Hello All,

I am following up on a discussion in the tracking software meeting this morning. This email is intended to be a conversation starter.

Although MDC 2 is 6 months away, we have to start planning now for what we will attempt to accomplish on the tracking side. Broadly, we need to establish:

 1) What tracking capabilities do we intend to exercise in MDC2?

 2) What do the analyzers hope to get out of it?

 3) How will we evaluate the physics performance?

1) Tracking capabilities:

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

I think we should try to make MDC2 look as much as possible like the analysis of real data. Namely:

All simulations done with

 - Full event pileup

 - De-aligned and de-calibrated detectors

 - Space charge distortions present

 - Central membrane laser flash hits

All calibrations derived from simulated data

 - Detector alignment

 - Gain calibrations

 - Space charge corrections

Calibrations applied during reconstruction

 - Detector alignment

 - Gain calibrations

 - Space charge corrections

Complete track reconstruction without the use of any truth information

 - For tracks from the primary vertex

 - For secondary tracks

Capabilities not presently existing, or not under development, are shown in bold. Reconstruction of secondary tracks is included among those because the tracking is not optimized to handle them when doing reconstruction with no truth information . Optimizing the treatment of secondary tracks is a project for the near term. We presently have no provision in the tracking for alignment or gain calibrations, or for simulating de-aligned and de-calibrated data. Input from the calibration task force would be very helpful there. Note that full event pileup applies for p+p collisions also!

2) What do analyzers hope to get out of it?

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

We should understand what plans or goals the analyzers have for analyzing MDC2 data, so that we can try to make that feasible. What do you do now, and what needs to change so that you can extract physics from DST's that were produced with no truth input at all? Discussions with the topical groups will be needed for that.

3) Evaluation of performance

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

The simulation part of MDC2 will have to produce

 - Simulated g4 hits

 - Simulated reconstructed hits

These will be in separate files.

To evaluate the reconstruction results, we will have to read in the g4 hits file and the reconstructed DST file.  But the existing tracking evaluation is quite slow compared with tracking. If we run evaluation on all events, it will dominate our resource use. We have to discuss what level of track evaluation the physics groups will need.

Perhaps we can touch on some of these topics during the collaboration meeting.

Sorry about the long email!

Tony