---

From Thursday's OffSoft meeting...  11/12/15

FTS Tracking
Hit errors were not transformed properly from radial to carteisian coordinates.  As a result, hit errros were larger than they should have been.   Errors are now correctly defined, and the vertex of (0,0,0) added for fits to primary tracks.  Both TPC tracking and FTS tracking (as 2nd pass) are enabled, so that we can run realistic events (e.g. hijing), reco the vertex with the TPC, and then include vertex as a constraint to the FTS tracks.

100 events (1 track/event) sample looks reasonable, though pT distribution appears somewhat non-statistical.  Action items: (1) run more events; (2) run more tracks / event.

StEvent in CVS has been adapted, along with MuDst, to include FTS. 

---

100 events
100 pi- per event
pT = 0.2 GeV
2.5 < eta < 4

R=64, PHI=12*128



With recent changes
(1) Efficiency for 100 tracks / event is around 96.5% for primary tracks. 
(2) pT reconstruction is not the best, though this is for *global* tracks.  Could be improved with primary vtx.

---

100 events
100 pi- per event
pT = 1.0 GeV
2.5 < eta < 4

R=64, PHI=12*128

---

Where does the Al part of the beam pipe come into play?  Beyond the nominal acceptance of the detector (eta > 4.0)