GSI Forum: Tracking » Eta

Home » PANDA » PandaRoot » Tracking » Eta_c efficiency

Show: Today's Messages :: Show Polls :: Message Navigator

Eta_c efficiency [message #12809]

Fri, 25 November 2011 19:20

Gianluigi Boca
Messages: 177
Registered: March 2004

first-grade participant

From: *gsi.de

Hi,
I modified again PndSttMvdTracking.cxx in order to get the 'old'
efficiency level, as in July11 release, for the Eta_c channel.
The difference, as anticipated, was a stricter cut on the association of the Parallel STT hits.
Please analyzers update and try again at your convenience

Tschuess Gianluigi

Report message to a moderator

Re: Eta_c efficiency [message #12810 is a reply to message #12809]

Sun, 27 November 2011 23:27

Dima Melnychuk
Messages: 213
Registered: April 2004
Location: National Centre for Nucle...

first-grade participant

From: *play-internet.pl

Hi,
I just want to quote two numbers concerning eta_c reconstruction efficiency I obtained with 1000 events statistics.
Before the last modification I obtained 28.4% after it 30.9%, so indeed it improves additionally reconstruction efficiency.
Both numbers are for the same data set.

It's still a question how those numbers changes with higher statistics.

Dima

Report message to a moderator

Re: Eta_c efficiency [message #12812 is a reply to message #12809]

Mon, 28 November 2011 12:43

Dima Melnychuk
Messages: 213
Registered: April 2004
Location: National Centre for Nucle...

first-grade participant

From: *fuw.edu.pl

Hi,

Just an additional number for eta_c efficiency.

For the data produced on grid without latest update I have eta_c reconstruction efficiency 23.5% with analysed 70.000 events.

Somehow with the grid data I have efficiency lower than with data produced by myself (28.4%). And it's not the first time I observe this behaviour.

Dima

Report message to a moderator

Re: Eta_c efficiency [message #12816 is a reply to message #12812]

Mon, 28 November 2011 16:05

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: *30-79-r.retail.telecomitalia.it

Could you please check if some subjob has 0 entries in the tree? Maybe there are some empty files, maybe...

Report message to a moderator

Re: Eta_c efficiency [message #12817 is a reply to message #12816]

Mon, 28 November 2011 17:14

Dima Melnychuk
Messages: 213
Registered: April 2004
Location: National Centre for Nucle...

first-grade participant

From: *fuw.edu.pl

Event if there are files which has 0 entries in the tree it would not change the efficiency which I quote since I normalise for the actual number of entries, at least if it is obtained correctly from:

PndEventReader evr(inPidFile);
nevts=evr.GetEntries();

Report message to a moderator

Re: Eta_c efficiency [message #12818 is a reply to message #12817]

Mon, 28 November 2011 17:43

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: *30-79-r.retail.telecomitalia.it

Considering that the efficiency is something like 20%, there should be no empty files. In such a case, maybe the reconstruction macros had some problems and have created a tree with no entry, without giving an error. The subjob is simply failing and it should not be counted in the 70000 statistics.
But I don't know if this can happen.

Report message to a moderator

Re: Eta_c efficiency [message #12827 is a reply to message #12809]

Tue, 29 November 2011 14:43

Dima Melnychuk
Messages: 213
Registered: April 2004
Location: National Centre for Nucle...

first-grade participant

From: *fuw.edu.pl

Hi,

With the latest STT reconstruction update the following results are obtained for eta_c channel with data produced on grid.

run926 (99500 events, copy of one file (500 events) fails)
Invariant mass distribution:
index.php?t=getfile&id=6732&private=0

Reconstruction efficiency - 25.8%, result is good but again somehow lower than I obtained with 1000 events (30.9%)
sigma(eta_c)=31.6 MeV, sigma(phi)=3.87 MeV

With cleanup - run926cu
index.php?t=getfile&id=6733&private=0

Efficiency drops to 18.1%, but still is reasonable.
sigma(eta_c)=28.7 MeV, sigma(phi)=3.70 MeV

And here are the multiplicities of charged tracks, without and with cleanup:
index.php?t=getfile&id=6734&private=0