Genfit with Forward Tracks [message #13400] |
Tue, 24 April 2012 18:08 |
StefanoSpataro
Messages: 2736 Registered: June 2005 Location: Torino
|
first-grade participant |
From: *36-79-r.retail.telecomitalia.it
|
|
Dear all,
let me announce that, after spending some time on the subject and thanks to several fixes in the fts geometry by Isabella, now genfit with forward tracks seem to work.
This means that, if you attach the kalman task after the FTS ideal tracking:
PndFtsTrackerIdeal* trackFts = new PndFtsTrackerIdeal();
trackFts->SetRelativeMomentumSmearing(0.02);
trackFts->SetVertexSmearing(0.02, 0.02, 0.02);
trackFts->SetTrackingEfficiency(1.);
trackFts->SetTrackOutput("FTSTrkIdeal");
fRun->AddTask(trackFts);
PndRecoKalmanTask* recoKalman2 = new PndRecoKalmanTask();
recoKalman2->SetTrackInBranchName("FTSTrkIdeal");
recoKalman2->SetTrackOutBranchName("FTSTrkIdealGen");
recoKalman2->SetBusyCut(50); // CHECK to be tuned
fRun->AddTask(recoKalman2);
you can have refitted tracks.
I have tried with 1000 muons at 2GeV/C (maximum dipole field):
You can see the momentum distribution, in blue after the ideal tracking setting a p smearing of 2% (of course the distribution has a width of 2%), and in red after the kalman task, obtaining a new momentum resolution of 0.6%.
Of course many things have to be checked, but I think this is a good starting point for this who would like to spend some time in forward tracking. Once everything will be fine I will update also the default macros in macros/pid, for the moment you have to use the code I wrote before.
Some feedback is welcome!
-
Attachment: fts_comp.gif
(Size: 7.65KB, Downloaded 539 times)
|
|
|