Hi Flemming, Djam, Peter, at al, [Shifted to from brahms-soft-l to brahms-dev-l - Please guys, annoucements on soft-l and discussion on dev-l - no need to bother everyone with various technical "issues"] On Fri, 2 Nov 2001 12:22:13 +0100 (CET) Djamel Ouerdane <ouerdane@nbi.dk> wrote concerning "new brat version 2.1.20": > Hi all, > > Since I'm on night shift but don't need to be focused on the > machines (cf latest shift report), I've added a couple of new things > in brat, therefore tagged it BRAT-2-1-20, and bumped the version > number to 2.1.20 > > Changes: > > added new class BrVertexCalibration, that deals with the offsets. I believe that's something we need. Looking at last years data (since there's still no good calibrations for the TMA and SMA for this years data), I found that if I used the ZDC and BB verticies, I got an extremly odd dN/deta for the SMA when compared to the same in the TMA. Ignoring the ZDC vertex solved this problem! > The modules/calib dir has a new subdir: vertex > The classes are (so far): > BrVertexCalModule (base class) > BrBbVertexCalModule > > (the TPC and ZDC are not yet available) Are they done and you're just holding back, or there's absolutely nothing yet? > > The scripts/calib/bb dir. has 2 new scripts: > BbVtxOffset.C > CommitVtxOffset.C Please add calib to SUBDIRS in scripts/Makefile.am, and add scripts/calib/Makefile.am (and so on) which installs the scripts into $(BRATSCRIPTDIR). See "The Guide" and scripts/Makefile.am for further references. > There usage is very simple but requires you have the latest > BrBbCalHitsModule and BrBbVertexModule that I've updated to deal > with the calibration (minor changes but necessary to be able to do > it). Did you intent to substitute the old BrBbRdoModule with these class or what? If so, why don't you rename BrCalHitsModule to BrBbRdoModule an BrCalHits to BrBbRdo? Also, did you make BrBbVertex derive from BrVertex (possibly defining a new class BrTpmVertex deriving from BrVertex to)? > I had to update configure.in and config/config.mk so that the > makefile sare generated and so on. What a surprise - there's nothing odd in this - see "The Guide". > I've tested it and it gave me very nice results. I inserted into the > sql DB as a test the BB offset values for run 5036. What is "very nice results"? > There are still a couple of things to update but it will come soon > (since I won't have many shifts now). Eagerly awaiting that code. > I must confess one thing: I couldn't test the thing from a new > version of brat since my own brat has a lot of new things that will > be committed this week-end (tracking thing and tof cal.) > > So, if you check out a new brat and can not compile and run the stuff, I > am the one to blame and I'll fix your problem asap. See also below. On Fri, 2 Nov 2001 07:04:35 -0500 "Flemming Videbaek" <videbaek@sgs1.hirg.bnl.gov> wrote concerning "Re: new brat version 2.1.20": > Djamel, > > We need this, but also a breif discussion on this. What better discussion basis than some code? > On the ZDC you need two offsets one for the 'regular > discrimintarors' and one for the offsets determined by the constant > fraction discriminators. > > Possible you also want to store the deduced values of sigma's > between TPM1, BB, ZDC,.. since these values have to be used in > analysis later for data selections > Since you admit beeing ignorant maybe you should have waited > createing the entry - but since done please change once we have > agreed on the content. As you perfectly well know, the revisions does not have a fixed size. In fact, it's designed so that one may change the number of parameters stored in a revision if the need arises (More slats in the TOFW is a good example). Hence, in so far as one calibration parameter (set) is all that's needed, there's little danger of making bad changes. But, a "Intent To Change" (ITC) message would have been appropiate. On Fri, 2 Nov 2001 13:12:06 +0100 (CET) "Peter H. L. Christiansen" <pchristi@nbi.dk> wrote concerning "Re: new brat version 2.1.20": > BRAT DOES NOT COMPILE!! > > modules/calib/vertex is empty but included in modules/calib/Makefile.am Did you remember to do cvs update -P -d You always need that when new directories are added to the CVS repository, hence I always use the -P and -d options to update. Yours, Christian Holm Christensen ------------------------------------------- Address: Sankt Hansgade 23, 1. th. Phone: (+45) 35 35 96 91 DK-2200 Copenhagen N Cell: (+45) 28 82 16 23 Denmark Office: (+45) 353 25 305 Email: cholm@nbi.dk Web: www.nbi.dk/~cholm
This archive was generated by hypermail 2b30 : Fri Nov 02 2001 - 07:43:54 EST