[Halld-cal] updated hdgeant/4 shower energy study

[Richard Jones]

Dear colleagues,

I redid the fits to the hdgeant/4 shower energy recons vs generated for
single-shower events in the fcal and bcal. This time, instead of just
taking the mean of the reconstructed shower energy, I formed a spectrum for
each bin in Egen and fitted it to a Gaussian, taking the mean of the
Gaussian fit. The results are shown in the latest update to the slides on
docdb.

https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=3852

The errors in these centroid values is smaller than the original TProfile
means that I had shown earlier because those were sensitive to a few events
in the low-energy tail. The new results do not show the polar angle
dependence that was seen using the original method. They do show the
negative offset that was seen before, and now it is a more robust result,
independent of whether the fit is first-order or second-order.

There is a small difference visible between hdgeant and hdgeant4 in the
slopes. I will take that out by tweaking the simulation once we have agreed
on an overall strategy.

-Richard Jones
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-cal/attachments/20190119/d2dad6a8/attachment.html>