[Sbs_gems] [Revised Logentry] SBS GEM tracking disabled from online replays for now

[puckett at jlab.org]

Logentry Text:
--
I was looking at the SBS GEM tracking results a little more carefully and at the moment I see no evidence that ANY of the reconstructed "tracks" in the SBS GEMs are real. The occupancies are VERY high and the tracking is merely fitting a lot of straight lines to noise. As such, the reported "efficiencies" are meaningless. To get meaningul SBS tracking results, the database needs to be tuned up using low-current data, and probably some code improvements are needed to implement more stringent search region constraints using the SBS magnet optics and possibly also BigBite vertex information. Moreover, including the SBS GEM tracking in the replay slows it down considerably. Since it is currently a waste of CPU cycles to attempt tracking in SBS GEMs at all, I have disabled SBS tracking from the replay by setting the "minimum hits on track" to 8 in the replay database. This means we will still decode the SBS GEM data to monitor hit rates/occupancies/etc, but for the time being we !
 will no longer attempt tracking until expert attention is given to making it work. 



---

This is a plain text email for clients that cannot display HTML.  The full logentry can be found online at https://logbooks.jlab.org/entry/4194870
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/sbs_gems/attachments/20231007/f69d266c/attachment.html>