Hi, I improved T4 local tracking in run06 where 1/2 of the 3rd module was dead. The changes are related to BrDCClusterModule by adding to lines if(!strcmp("T4",GetName()) && fRunNo>16000) DCSingle(0,3,2); // Plane 1-Y2 if(!strcmp("T4",GetName()) && fRunNo>16000) DCSingle(2,0,1); // Plane 3-X1 DCSingle can create "combined hits" for single plane only. Normally DCPair(T3/T4/T5) and DCTrio (T3) are used. By calling DCSingle(0,3,2) we enhanced missing region in Y (in this region we have y info from 3 y planes only as compare regular 4 planes). DCSingle(2,0,1) make the overall enhancement allowing only one working x plane in 3-rd module to be used in 2D track creation. I make a test for 2 sequences run 16199 (2A1723) and got the following results old ltr version new ltr version reco tracks 6795 7870 sigma 0.00908 0.00916 hits per track 19.14 18.99 My calculation shown that T4 efficiency with the old version was about 80%, so one could expect something like 93% with the new code. If /afs is back I will commit the updates and tag brat with the new version. Best regards, Pawel. -- ---------------------------------------------------------------------- | Pawel Staszel | | Jagiellonian University | | Institute of Physics email: ufstasze_at_if.uj.edu.pl | | Reymonta 4 phone: (+48) 12 663 5705 | | Krakow 30059 FAX: (+48) 12 633 7086 | | Poland | ---------------------------------------------------------------------- _______________________________________________ Brahms-dev-l mailing list Brahms-dev-l_at_lists.bnl.gov http://lists.bnl.gov/mailman/listinfo/brahms-dev-lReceived on Fri Sep 29 2006 - 09:37:43 EDT
This archive was generated by hypermail 2.2.0 : Fri Sep 29 2006 - 09:38:08 EDT