Hi Selemon, On the T4; It is likely that I only reran the 2 and 4 deg settings in question on t4; I actually checked the only run that were redone on Oct 30 after Pawel changed the DC code was 14392-14425 which is precisely right after the runs you question. By the way you plots also show low effeciency for 8B2442 8B1723 so probably you should rerun all of 14339-14391 ltr,gtr and dst. On h1/h2 efficiency; As pawel say this cannot be used, acouple of reasons is i) it is not done per pointed slat, so dead slats are not accounted for properly. ii) matching etc may not be done precisely as in Brat as pawel points out. So the proper thing is to evaluate it with the dst. It is basically very high once you take out the deadslats. flemming -------------------------------------------- Flemming Videbaek Physics Department Bldg 510-D Brookhaven National Laboratory Upton, NY11973 phone: 631-344-4106 cell: 631-681-1596 fax: 631-344-1334 e-mail: videbaek @ bnl gov ----- Original Message ----- From: Pawel Staszel To: Bekele, Selemon ; brahms-dev-l_at_lists.bnl.gov Sent: Thursday, January 25, 2007 5:00 AM Subject: Re: [Brahms-dev-l] Tracking efficiency in run05 pp Hi Selemon, I'm glad that it works. If you have found that some important information is missing in the README file, fill free to update the file. Concerning T4 (B setting, runs: 14339-14348): the latest ltr production was done on 2005-09-24. We discovered the T4 efficeincy problem (on last year) looking at other settings and the DC tracking code was updated on 2006-10-27 to enhance the efficiencies in T4. I think that after that Flemming re-run ltrs for runs showing the problem but probably not for all of them. I suggest you to redo ltrs for one run (e.g. 14339) and see whether new T4 efficiency is reasonable (on the level of 95%), if so, redo the whole set. Concerning H2/H1 efficiencies, Although efficiency files contain maps for H1 and H2 one should not rely on them, because the track versus tof hit matching criteria used in the efficiency stuff are not identical as these applied in the reconstruction chain. Once I tried to fixed it but could not easily understand of how this matching is done in the brat. But it fact what we need for H1/H2 is the final pid efficiency (=efficiency that the hit was detected, matched to track and then successfully used to identify the track). We can retrieve this final pid efficiencies in the similar way as we do it for Rich (e.g. from the dst files). Maybe I will show the example of such analysis on the coming phone meeting. Cheers, Pawel. Bekele, Selemon wrote: Hi Pawel, I managed to make the efficiency files finally. Please have a look at the figures in http://www4.rcf.bnl.gov/~tigist/EfficiencyPlots/run05_pp/ I am worried about the efficiencies for T4 (B setting) T22 H2 May be increasing the sigma cut would improve things. Is this something you noticed while making the efficiencies at 4 degree? Selemon, -----Original Message----- From: Pawel Staszel [mailto:ufstasze_at_if.uj.edu.pl] Sent: Tue 1/23/2007 3:38 AM To: Bekele, Selemon Cc: brahms-dev-l_at_lists.bnl.gov Subject: Re: Tracking efficiency in run05 pp Hi Selemon, My feeling was that it should run anyhow, even if Enc offset files are not there. It usually means that enhancement is not really necessary which is quite likely for pp data. I will look at the problem for run 14274. Cheers, Pawel. Bekele, Selemon wrote: Hi Pawel, thanks for your reply. I found in the files you made effiencies for the 4 deg data for run05 pp. I have been trying to make efficiency files for run05 pp at 8 degrees. I am using gtr production 4 and some gtr DC enhancement offset files are missing for some of the runs. For example, I see run 14351 offset014351_seq002_seq003_seq004_seq005v0p4.off.EncT3 OK offset014351_seq002_seq003_seq004_seq005v0p4.off.EncT5 OK run 14322 offset014322_seq002_seq003_seq004_seq005v0p4.off.EncT3 not there offset014322_seq002_seq003_seq004_seq005v0p4.off.EncT5 14274 *.EncT3 not there *.EncT5 not there I am running into problems while submitting jobs automatically. I am not sure why these files are not generated for some runs when the gtr files are produced. Selemon, -----Original Message----- From: Pawel Staszel [mailto:ufstasze_at_if.uj.edu.pl] Sent: Fri 1/19/2007 3:11 AM To: Bekele, Selemon Subject: Re: Tracking efficiency in run05 pp Dear Selemon, Bekele, Selemon wrote: Hi Pavel, hope you are doing fine. Thanks, I'm fine. I am trying to make efficiency files for run05 pp. I am not sure if there are files already made for all settings in the FS. Would you please let me know if that is the case. I made efficiency calculation for run05_pp in August06. The maps are at the official location e.g. /brahms/data08/corrections/trackEff/MappedEffic_run05_pp_200_PSv1.root To check out whether the settings you need are there you can just edit the root file in the root session and do >.ls, histogram's names contain information about the settings. If not, I would need to generate them myself and it would be great if you already have scripts ready to run automatically. I know there would be no centrality issue here, but I expect efficiency as a function of the "X position" and momentum. The stuff used to generate maps (../banapp/corrections/efficiency) includes README file which provides step by step instruction of how to generate the maps. Concerning variables: it is your choice which variables to map. When running MapEfficiency.C script (the last stage of the process) you have to specify string VarRequested like "XAP" (Length has to be equal 3). In this case you will create X vs X-slope 2D map and 3 1D maps - for X. X-slope and momentum (for each detector). If VarRequested="XPA" then the 2D map will be X vs momentum. To use the maps ones use EfficiencyCalculator (which is fact is efficiency interface). EfficiencyCalculation has setters void SetVar(TString var). if var is "XA" the EfficiencyCalculation will try to use 2D X vs X-slope maps. If var is "X" or "P" it will try to use 1D X map or 1D momentum map. If the maps will not be found (it refers only to particular settings) the calculation will inform about it and abort. If you have more questions, then let me know. Cheers, Pawel. Thanks for the help, Selemon -- ---------------------------------------------------------------------- | 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-l _______________________________________________ Brahms-dev-l mailing list Brahms-dev-l_at_lists.bnl.gov http://lists.bnl.gov/mailman/listinfo/brahms-dev-lReceived on Thu Jan 25 2007 - 07:04:42 EST
This archive was generated by hypermail 2.2.0 : Thu Jan 25 2007 - 07:05:06 EST