Hi, Introduction: the EfficiencyFinderModule is used on the event by event basis but the gtr offsets script produces parameter files by merging usually two (sometimes more) sequences. Up to now Efficiency Finder was using parameters calculated for the two first sequences, and I thought that it is OK. However digging in some other things I realized that these parameter can differ significantly from sequence to sequence, e.g. T2T3Offsets: seq0_seq1 seq006_seq007_seq008_seq009 Dx -0.102672 +/- 0.0703688 -0.0967937 +/- 0.0779666 Dy 0.038688 +/- 0.139611 0.036911 +/- 0.15073 Dax -0.0001 +/- 0.000739 -7.19848e-05 +/- 0.000931802 Day 0.000243 +/- 0.000607 0.00026091 +/- 0.00101402 As you see deviations are quite large, in case of Day we have deviation in sigma on the level of 40%. (You can find documentation of this effect on http://zefir.if.uj.edu.pl/staszel/brahms/fseffic/seqbyseq/index.html) Because of that problem I updated EfficiencyCalculator (banapp) in the way that in can now update gtr offsets on the sequence by sequence basis. It was not as trivial because our modules do not have seq. by seq. functionality. It required some changes to BrModuleMatchTrack, BrDcEnhancementModule and BrT2T3OffetModule, mainly, I moved the section that reads offsets from Init to separate public method ReadOffsets. Is is now called from the Init but also can be accessed from EfficiencyFinderModule. The changes do not require modification of the official reconstruction scripts. New brat tag is 2.35.2 Regards, Pawel. -- ---------------------------------------------------------------------- | Pawel Staszel | | Jagiellonian University | | Institute of Physics email: ufstasze@if.uj.edu.pl | | Reymonta 4 phone: (+48) 12 663 5712 | | Krakow 30059 FAX: (+48) 12 633 7086 | | Poland | ---------------------------------------------------------------------- _______________________________________________ Brahms-dev-l mailing list Brahms-dev-l@lists.bnl.gov http://lists.bnl.gov/mailman/listinfo/brahms-dev-lReceived on Wed May 24 08:47:53 2006
This archive was generated by hypermail 2.1.8 : Wed May 24 2006 - 08:48:06 EDT