On Fri, 24 Aug 2001, Christian Holm Christensen wrote: > > I've made a module for processing TPC pedestal data. Before checking > > it into BRAT I should ask if anyone has any opinion as to where this > > module should be placed. I wonder if the right place should be a > > new <brat source dir>/modules/calib/tpc. If I don't hear/see any > > objections I'll put it there in a few hours. > > If BrTpcPedestalModule is a module that makes some sort of > calibrations (pedestals for the TPCs? Is there not supposed the be > zero suppression in the DAQ? Are the pedestals really used in the > reconstruction?) then it should go into modules/calib/tpc AND is > should be called BrTpcPedestalCalModule (or shorter > BrTpcPedCalModule). Sure there is both zero subtraction and suppression in the DAQ. However, this work is dependent on the quality of the pedestal data taken in pedestal runs. As the TPCs have shown to give worse pedestals this year than last year, mainly due to the fact that we are now pulsing the gating grids on all TPCs, the pedestals have to be monitored. Since the output of this module is not going to be used for calculating pedestals to be sent to the DAQ, I guess that my suggested location for the module is not reasonable after all. > I'd like to take this oppertuinity to point out that "The Guide" > actually describes this. Perhaps someone at BNL should print 20 > copies and glue it onto the pii machines, preferably over the monitor > so that you cannot help see it :-). How about desktop background and screensaver versions? :) > If it's not used in the tracking/analysis or doesn't make > calibrations, then it does not belong in BRAT. Keep it in brahms_app > in that case. Well, the act of investigating pedestal data I would absolutely consider to be analysis. However, the idea is that this modules and a program making use of it should be used by the person(s) on shift to monitor the quality of the TPC pedestals. So it may not be of such a general character to put it into BRAT, at least not yet. I'll put into brop together with the analysis program. > Keep in mind, that everytime you add a class to BRAT, > the "confusion factor" goes up one notch. If the class isn't used, > then the confusion factor goes up two notches. Well, as a wise man once said, if the class has a well written description including typical use and also the comment after ClassDef call is well chosen, the class will contribute less than 10% of a notch. That is really worth keeping in mind. JI -- Jens Ivar Jordre e-mail: JensIvar.Jordre@fi.uib.no usually: Dep. of Phys., Allegt. 55, N-5007 BERGEN, NORWAY currently: Bldg 510D, P.O.Box 5000, Upton, NY 11973-5000, USA phone: +1-631-344-4223 --------End of Unsent Message
This archive was generated by hypermail 2b30 : Fri Aug 24 2001 - 05:54:42 EDT