ivana [Tue, 3 Nov 2009 10:54:49 +0000 (10:54 +0000)]
In reconstruction:
Add the possibility to adjust chamber by chamber the x/x0 value used
in the computation of Multiple Coulomb Scattering effets (driven by the fact
that it is ~2 times higher for St1&2 than for others).
The choice of these values has a little effect on the resolution of the
tracking, but it can change significantly the size of the area where
we search for new clusters to be attached to the track.
(Philippe P.)
ivana [Tue, 3 Nov 2009 10:42:26 +0000 (10:42 +0000)]
In trigger:
- Add possibility to fill the trigger board efficiency with the average
value of the corresponding RPC.
- Update MUONTriggerChamberEfficiency.C
- Update READMEtrigger.txt
(Diego)
morsch [Tue, 3 Nov 2009 10:00:14 +0000 (10:00 +0000)]
These are the changes for the propagation of the tag cut selection
bookkeeping.
I have tested it in the following way:
- Created XML list with cut summary in, read it via the
AliTagAnalysis, put it as a UserInfo object into the chain and recovered
the information in the task via the ESD Handler
- Checked that the same code does not crash if the tag summary
information is not available in the XML list.
richterm [Tue, 3 Nov 2009 07:39:41 +0000 (07:39 +0000)]
temporarily disabling the initilization of the Root Cint error callback. It expects a void*,
but the callback is of course a function*. On some systems, the warning is interpreted as
error. Have to check whether the definition of the fuction in Root needs to be changed to
void (*)() as argument, or whether there is another workaround.
sgorbuno [Mon, 2 Nov 2009 22:56:21 +0000 (22:56 +0000)]
SPD vertexer component created which reconstructs the XYZ primary vertex with SPD pixels.
The resolutions are about 400um.
It can treat an initial beam offset up to 1.5 cm in XY.
The component can do a calibration of the beam position for every N events
The speed on pp is ~5000Hz without producing histogramms and ~800 Hz with histograms
cblume [Mon, 2 Nov 2009 17:36:40 +0000 (17:36 +0000)]
Update by Theodor:
- clusterizer: AliInfo -> AliDebug for unimportant message
- tracker: decrease the multiplicity of an error message in AliTracker, which actually should not be an error message
kkanaki [Mon, 2 Nov 2009 15:51:36 +0000 (15:51 +0000)]
- use new function for calculation of track length
- implemented option -dEdx for plotting the respective histogram, default is kFALSE (no updates in the OCDB entry of the component)
sgorbuno [Mon, 2 Nov 2009 12:54:41 +0000 (12:54 +0000)]
- Method to calculate the track length is added to the AliHLTGlobalBarrelTrack
- Merged TPC tracks are not propagated to the TPC inner radius, but stays at the innermost cluster, in order to get the right track length
hristov [Mon, 2 Nov 2009 12:51:07 +0000 (12:51 +0000)]
Fixes in reconstruction:
1) Define in a single place (AliMUONTrack) the algorithm to be used to associate 2 tracks by comparing the position of their clusters. Update all the classes/macro using that algorithm to find the simulated track corresponding to a given reconstructed one:
AliMUONTrack:
- Change the function IsValid(...) to be able to check the reconstructibility of the track depending on the tracking algorithm
- Change the name of the function CompatibleTrack to FindCompatibleCluster that is actually what the function does
- New function Match(...) to associate, or not, 2 tracks by comparing the position of their clusters (can be used to assign MC labels to reconstructed tracks).
AliMUONRecoCheck:
- Check the track reconstructibility according to given informations about the tracking algorithm (those info can be extracted from RecoParam).
- New function FindCompatibleTrack(...) to find in the given store a track that match the given track. Matching is done by using MC labels or by comparing clusters' position.
All other classes/macros/tasks:
- use these new functionnalities
2) Additional correction:
DecodeRecoCocktail.C:
- correct the algorithm to make it working even if there are ghost tracks
- avoid accessing OCDB when not needed
MUONRecoCheck.C:
- new histograms/graphs to study the momentum resolution as a function of the momentum
- new histograms/graphs to study the cluster resolution at each chamber
- avoid loading magnetic field when not needed
AliMUONTrackLight:
- correct constructor
- avoid using magnetic field when not needed
marian [Mon, 2 Nov 2009 12:04:26 +0000 (12:04 +0000)]
AliTPCcalibDB.cxx - Calculate the distance to the closest measurement
AliTPCcalibDButil.h AliTPCcalibDButil.cxx - graph filtering functions
- calculation of drift velocity CE
CalibMacros/CalibEnv.C - dump all drift velocity option
misko [Mon, 2 Nov 2009 07:36:57 +0000 (07:36 +0000)]
GetOutputData() now used in AliAnalysisTaskUnicor::Terminate (thanks
Mihaela).
Using integers in AliUnicorHN::Fill was leading to trouble (thanks
Matevz, via his valgrind output), now all double.
Two track cuts modified, a monitoring histogram added.
Trigger phi was being overwritten in AliUnicorAnalHighpt, fixed.
Merge method added in AliUnicorAnal so one can run on proof.
Some antiviolence (with respect to coding rules) measures taken.
aszostak [Sun, 1 Nov 2009 18:01:41 +0000 (18:01 +0000)]
Fixing bug in AliHLTRunSummary and AliHLTEventSummary, need to have version number in CalssDef non-zero to propagate the objects through an HLT chain properly.
Fixing bug in AliHLTGlobalTriggerComponent to generate correctly trigger menu evaluation logic for trailing operators.
Fixing AliHLTGlobalTriggerWrapper to catch CINT errors correctly.
Fixing bug removing registrations in AliHLTConfiguration.
Adding more documentation.
mtadel [Sat, 31 Oct 2009 16:14:01 +0000 (16:14 +0000)]
* AliEveEMCALSModule
- Remove palette management that was cut-n-pasted from TPC code and
was not used in EMCAL code at all.
- Comment out manual destruction of (shared!) reference counted objects.
This needs to be improved further as static objects seem to be used
in a wrong way.
* AliEveEMCALData
- Reduce debug output, set debug to 0 by default.
- Do not delete AliEMCAL object - it is deleted by run-loader.