jgrosseo [Tue, 18 Oct 2011 10:20:15 +0000 (10:20 +0000)]
1. Update to AliESDtrack
- Adding "golden TPC cut" on chi2 between tpc only track constrained
to the vertex and the global track. The code is an adapted version
from the implemention provided by the RAA group (Jacek et al).
- The calculated chi2 value is cached to avoid the computing-intensive
operation several times in train running.
- New function that returns the number of crossed rows and caches it
(suggestion by Jens W).
2. AliESDtrackCuts has been optimized
- Golden cut added. Therefore the prototype of AcceptTrack needed to
be changed (backward compatible). One needs to pass the event as well
when the new cut is to be used.
- CPU intensive cuts are only evaluated when other cuts have
passed (Nsigma, golden TPC cut).
NOTE: Control histograms are only filled for those quantities when the other cuts have passed
jotwinow [Tue, 18 Oct 2011 09:26:47 +0000 (09:26 +0000)]
- changes by Markus Koehler
- histogram ranges for high pT and new varible (TPC found/findlable clusters) added
M TPC/AliPerformanceDEdx.h
M TPC/AliPerformanceDEdx.cxx
- histogram ranges for high pT:
M TPC/AliPerformanceTPC.cxx
M TPC/AliPerformanceEff.cxx
M TPC/AliPerformanceDCA.cxx
M TPC/AliPerformanceMatch.cxx
M TPC/AliPerformanceMC.cxx
M TPC/AliPerformanceRes.cxx
richterm [Tue, 18 Oct 2011 09:16:52 +0000 (09:16 +0000)]
using TObjString::String instead of GetString to get directly to reference to TString object and avoid temporary instance in return; takes account for changes in the bookeeping of TString objects in root v5-30
richterm [Tue, 18 Oct 2011 09:16:07 +0000 (09:16 +0000)]
using TObjString::String instead of GetString to get directly to reference to TString object and avoid temporary instance in return; takes account for changes in the bookeeping of TString objects in root v5-30
gconesab [Sun, 16 Oct 2011 19:04:50 +0000 (19:04 +0000)]
Add more shower shape long axis histograms for different weights, comment the ones for short axis
QA: add histograms related to energy around cell, for exotic cells studies
Electron: Shower shape weight histograms initialized twice, now corrected
gconesab [Sun, 16 Oct 2011 17:42:44 +0000 (17:42 +0000)]
Add cell rejection based on energy in sorrounding cells, 4 cells in corners with common side, a la CMS, if too low energy, and not noisy by time cut, reject cell before clusterizing, candidate of spike/exotic cell
Move parts of code in UserExec to independendt methods, to improve readability.
Order methods in source code in alphabetical order.
pchrist [Fri, 14 Oct 2011 11:37:38 +0000 (11:37 +0000)]
- Removed small bug (otherwise chargeVector was not filled for normal data)
- Added Additional track cuts (for further systematic studies)
- Consistency between ESD and AOD analysis
agheata [Fri, 14 Oct 2011 08:37:28 +0000 (08:37 +0000)]
SDDRP - removed cdb->SetRun, initialize CDB clib objects at the first call of UserExec rather than UserCreateOutputObjects. ITStracking, ImpParRes - added PostData in UserCreateOutputObjects. SPD - no need to load geometry since in the train it is already done by AliTaskCDBconnect (protected via fTest)
richterm [Thu, 13 Oct 2011 23:47:04 +0000 (23:47 +0000)]
adding publishing of histograms, by default all histograms are sent as separate objects; option '-publishing-mode off/separate/list/array' to switch publishing off or publish everything in a TList/TObjArray; removed the default histogram file name, so nothing is stored in local files unless the option '-histogram-file' is used to configure that feature
richterm [Thu, 13 Oct 2011 20:35:01 +0000 (20:35 +0000)]
adding further optimized compression format with additional 15% compression ratio in normal compression mode; correcting rounding errors in track model compression to make it lossless with respect to normal mode
cvetan [Thu, 13 Oct 2011 07:15:14 +0000 (07:15 +0000)]
Since the trigger bits are not available in the V0 raw-data header, we replace the trigger selection by a cut on the minimum number of TDC flags present.
gconesab [Wed, 12 Oct 2011 16:50:36 +0000 (16:50 +0000)]
Correct method combining conversion photons and calorimeter clusters, now fill also shower shape and weight histograms for the clusters under neutral meson peak