[Brahms-dev-l] new Fs efficiency package

Dear Colleagues,
Let me announce the release of the new reconstruction efficiency
package. It can be check out from ps_app/fs/receffnew.
README file contains quite good (I guess) documentation.
The advantage of this new package over the old one from  ps_app/fs/receff
is due to the matching criteria: The new module uses the same criteria
for matching as these used in BrModuleMatchTrack by calling:
Float_t dist = fModuleMatchTrack->MatchFrontBack(frontTrack,backTrack,p);
In means that we refer directly to the BrModuleMatchTrack to get
the track matching parameter (this is distance in 3D space of dy, dang
and daly, see BrModuleMatchTrack::MatchesCuts and
BrModuleMatchTrack::MatchingParameter).
MatchFrontBack is a new public function of BrModuleMatchTracks that does 
the similar job
as CombineFrontBack but acts only on two (front and back) tracks and not 
on the two track tables.
(I don't think it is a perfect solution but for now it is ok). New brat 
tag is 2.24.6.

This new approach solves the problem of the efficiency dependency on the 
sigma
range parameter. Now, if the same sigma range in used in global tracking and
EfficiencyFinderModule the extracted efficiency (for the full FS) refers 
to the combined 3 important corrections:
detector efficiencies + local tracking efficiencies + global tracking 
efficiency. That's what we
need to get the absolute yields.
(To get the efficiency the all chambers (full FS) we call MatchFrontBack 
for T1-T2, T3-T4, T4-T5, and
T3-T2 matching is done as it was in the old package. The same matching 
is applied in the FsTrackingModule
combine the full Fs track!!!
Note, that the matching parameters used here are the same as these 
extracted and used in the official global tracking.

Still EfficiencyFinderModule has its own matching parameters but these 
are only used to confirm the
reference tracks, e.g. to confirm whether the  T3-T4 track is a good 
reference for the T5 efficiency we match it with
H2 hits, and to confirm whether T2-T3-T4-T5 is good for T1 we match it 
with vertex (at least for auau).


Regards
Pawel.



_______________________________________________
Brahms-dev-l mailing list
Brahms-dev-l@lists.bnl.gov
http://lists.bnl.gov/mailman/listinfo/brahms-dev-l