sPHENIX HCal discussion

[John Haggerty <haggerty AT bnl.gov>]

I was thinking about the plots we'll need to come out of the ongoing 
simulation workshop and the work that will follow it, and I was thinking 
that rather than defining all the plots we want right now, I'd suggest a 
way to preserve the information we want so that we can not only make the 
plots we want, but have a way to archive the results, update them, and 
create a uniform look to the plots.

I'm going to focus on the single particle studies for now, but we might 
be able to extend this to jet studies as well.

The basic idea is that we summarize the single particle studies in an 
ASCII file that can be imported into ROOT as a TTree or TGraph, an Excel 
spreadsheet, a database... whatever.  We characterize the distributions 
in a general way and store it in a row of the table.  We can then 
generate plots of anything we want pretty quickly, and be able to fix up 
axis labels and colors and whatnot without going back to the data.

Does that sound like a workable approach?

I probably haven't thought of everything, but the columns I can think of 
right now are (with some redundancy and some luxury items) are below (I 
would stick to numbers to make it easier to parse):

- run number (so we can go back to data if necessary)
- number of events
- B field
- active detector systems (I would use a bitmap, but whatever)
- If it's a variant on the reference design, we'd need a way to identify 
it, maybe a version number
- input particle type
- incoming angular range
- For each calorimeter section (EMCAL, IHCAL, OHCAL, and the combination?):
  o mean
  o sigma
  o Gaussian fit mean
  o Gaussian fit sigma
  o Characterization of low end and high end tail--I can think of 
several methods, but that needs some thought

I think that's enough to make the basic linearity and resolution plots, 
show what kinds of tails the distributions have, correct for sampling 
fraction in different ways and the other basic information we need to 
show, although I may have forgotten something, or perhaps someone sees a 
flaw, or even better, ways to extend it and make it more general.  Maybe 
we can have a small subcommittee to discuss this today while we are 
together.

--
John Haggerty
email: haggerty AT bnl.gov
cell: 631 741 3358