[Primexd] Data production

[Drew Smith]

Hi Sean,


>From my point of view, before calibrations can be finished we need to add a position correction (for both methods of calculating position) for the reconstructed showers. After that, it would be good to re-calibrate the gains, because I use Compton Scattering events to tune them. I don't want to underestimate how long this will take, but I will try to determine such a correction this week using data. However, it may be the case that we need to use MC to fine tune it.  After that, another pass of tuning the gain factors would be pretty quick.


I'm not sure if there are any other items that people would like to mention, but after this I think the calibrations would be in a fairly good state for this past run period. Of course there are many other things that will need to be studied and characterized in further detail (like dependence on incident angle, differences between electrons/photons, reconstruction of particles other than electrons/photons, and the dependence of the resolution on position within the calorimeter, just to name a few).


Best Regards,

-- Drew

________________________________
From: Primexd <primexd-bounces at jlab.org> on behalf of Alexander Somov <somov at jlab.org>
Sent: Saturday, July 20, 2019 9:48:11 AM
To: Sean Dobbs <sdobbs at fsu.edu>; primexd at jlab.org <primexd at jlab.org>
Subject: Re: [Primexd] Data production

Thanks Sean,

It would be great if you can help to
implement CCAL to the rest format.

We'll work on skims needed for data
analysis.


Cheers,
           Sasha
________________________________
From: Sean Dobbs <sdobbs at fsu.edu>
Sent: Friday, July 19, 2019 4:42:55 PM
To: primexd at jlab.org <primexd at jlab.org>; Alexander Somov <somov at jlab.org>
Subject: Data production

Hi all,

Sorry for not bringing this up at the Primex meeting - I got pulled
away in the middle and didn't get back until it was over.

There was one thing that I wanted to bring up:  in the Wednesday GlueX
Analysis meeting, the question was asked when the Primex data would be
ready for a first production pass to generate data in the standard
reconstruction format.  My take on this was that
1. The calibrations have to converge to a (not necessarily final)
state which everyone is happy with
2. Whatever skims would be helpful to generate for calibration and
data analysis should be defined and implemented

It would be good to have some clarity on where we stand with these two
items and what needs to be accomplished to complete them.  If anyone
else has some feedback on this subject, I'd be happy to hear it.

Once Drew's new version of the CCAL reconstruction is merged in, there
will be some minor changes to our REST format needed, which I can help
with.  I'll think if any other changes need to go in.

Cheers,
Sean
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/primexd/attachments/20190720/2873d7f3/attachment-0002.html>