[Brahms-dev-l] Suggestion for cleanup in the members in BDST/dst trees

From: Truls Martin Larsen <trulsml@nbi.dk>
Date: Thu Jan 05 2006 - 12:23:08 EST
Hi, 

due to all the resent changes in the members in the dsts I have some
more suggestion for cleaning up this, now quite big tree structure, a
little bit.

Here are some of branches:

G_fRunNumber : Is a copy of R_fRunNumber
G_fNoBfsTracks: should be replaced by a G_fNoFsTracks

Plus it would be nice to have the number of hits in the TPC's in the
tree:

G_fNoMrsHits
G_fNoFfsHits
(maybe also the G_fNoBfsHits???)

for the tracking efficiencies...

Also:

FS_fRich_fMass : change to fMass2 (which requires update of BrRichModule
to not do sqrt(abs(rich_mass_squared)) and only store the mass...

Also maybe slight code change in the way the number of triggers are
stored in the tree. Now they are just copied from the rundb. At a first
glance this might be fine, but if this number is used when doing
normalization one might do something wrong:

You have a big run of 15+ sequences. For some reason the daq kills the
run in the end by something like "iostream missing" (or something
similar, use your imagination from taking shifts). This will cause the
ltr production to not produce the last sequence in the run, though the
other sequences are fine. If one then make a dst of this bunch of
sequences there is nothing wrong with it, except that is has a nasty
trap if one accidentally normalizes with (G_fNTrigger4*G_fScaledown4).
(This can not happen unless you do a mean bias analysis though...)

Counting triggers and then storing makes it impossible to save it in the
tree if they are not counted and stored in the at a lower reconstruction
level...
Maybe G_fNTriggers should not be there at all(?) forcing a count of the
triggers upon analysis... And they are just a copy from the DB anyhow...

Just some thoughts...

I can do the changes if you agree to any of my suggestions...

Cheers,
	Truls
-- 
*---------------------------------*
|http://www.nbi.dk/~trulsml      /
|Truls Martin Larsen            /
|trulsml@nbi.dk                .
|The Niels Bohr Institute     //
|Work Address:               / \0
|Blegdamsvej 17             /\_/
|DK-2100 Copenhagen        /  /
|Tel: +45 35325269        / --
|                        /_/  |
|Home address:          /     \
|Holger Danskes Vej 28H|      '
|DK-2000 Frederiksberg |
|Denmark               |
|Mob: +45 20974802     |
*----------------------*


_______________________________________________
Brahms-dev-l mailing list
Brahms-dev-l@lists.bnl.gov
http://lists.bnl.gov/mailman/listinfo/brahms-dev-l
Received on Thu Jan 5 12:22:04 2006

This archive was generated by hypermail 2.1.8 : Thu Jan 05 2006 - 12:22:17 EST