GSI Forum: General » the eta_c analysis presented yesterday by Dima

Home » PANDA » PandaRoot » General » the eta_c analysis presented yesterday by Dima

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

the eta_c analysis presented yesterday by Dima [message #12428]

Tue, 02 August 2011 15:32

Gianluigi Boca
Messages: 177
Registered: March 2004

first-grade participant

From: *gsi.de

Hallo,
as some of you know, yesterday at the Tracking EVO meeting Dima presented the Eta_c analysis with the STT.

This analysis was performed on his computers with the latest version of the code in svn.

The shocking news was that the efficiency for STT dropped dramatically and the quality of the reconstructed Eta_c mass was essentially crap compared to the same analysis performed about 1 month ago in the Grid.

So last night I rerun all the efficiencies studies and I reproduced again all the efficiency Table that I presented many times in the past.

This table is a list of track reconstruction efficiencies and hit finding efficiencies and spurious percentages done with Box generated events of 1, 4, 10 tracks at 0.3, 2, 5 ant 10 GeV/c.

To my personal relief I noticed that nothing has changed WITH THE LATEST VESRION OF THE STT+MVD PATTERN RECOGNITION with respect to the past and all the efficiencies are at a very high level.

So I CLAIM THAT SOMETHING FISHY IS GOING ON IN THE CODE AFTER PATTERN RECOGNITION.

This of course is not Dima's fault. I invite every developer to investigate any change they may have done that affects the Eta_c results so much
thanks in advance Gianluigi

Report message to a moderator

Re: the eta_c analysis presented yesterday by Dima [message #12429 is a reply to message #12428]

Tue, 02 August 2011 15:44

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: 94.165.121*

Are you sure the macros are the same of the ones in grid? I amnot sure if the stt macros in svn were recently updated.

Report message to a moderator

Re: the eta_c analysis presented yesterday by Dima [message #12430 is a reply to message #12428]

Tue, 02 August 2011 16:00

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

first-grade participant

From: *fuw.edu.pl

Hi Stefano,

I compared macros I used locally for STT reconstruction with that on the grid and the only difference is latest update by Lia:

  PndMCTrackAssociator* trackMC0 = new PndMCTrackAssociator();
  trackMC0->SetTrackInBranchName("SttMvdTrack");
  trackMC0->SetTrackOutBranchName("SttMvdTrackID");
  trackMC0->SetPersistence(kFALSE);
  fRun->AddTask(trackMC0);

  PndSttMvdGemTracking * SttMvdGemTracking = new PndSttMvdGemTracking(0);
  SttMvdGemTracking->SetPdgFromMC();
  fRun->AddTask(SttMvdGemTracking);

plus added commented line by Gianluigi
// SttMvdTracking->Cleanup();

But I already tried with this line:
SttMvdGemTracking->SetPdgFromMC();
commented and it didn't cause any difference.

And in addition I just checked STT reconstruction with commented line
recoKalman->SetIdealHyp(kTRUE);

but difference is minor.

Dima

Report message to a moderator

Re: the eta_c analysis presented yesterday by Dima [message #12431 is a reply to message #12430]

Tue, 02 August 2011 16:07

Johan Messchendorp
Messages: 693
Registered: April 2007
Location: University of Groningen

first-grade participant

From: *Red-80-33-153.staticIP.rima-tde.net

A question to Dima: the loss of STT efficiency appears already in the most up-to-date july11 release? Or did it occur afterwards?

Greets,

Johan.

Report message to a moderator

Re: the eta_c analysis presented yesterday by Dima [message #12432 is a reply to message #12428]

Tue, 02 August 2011 16:16

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

first-grade participant

From: *fuw.edu.pl

Hi Johan,

With grid data produced with july11 release (run 921) results are good.

The problem exists with the trunk version as of July,31.

Dima

Report message to a moderator

Re: the eta_c analysis presented yesterday by Dima [message #12434 is a reply to message #12430]

Tue, 02 August 2011 16:38

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: 94.165.121*

I have not well understood. In the good macro there should be two mctrackassociator, one after the sttmvdgem task and another after the kalman. Moreover, inthe kalman one should set the pndtrackid TCA name. Was this done?

Report message to a moderator

Re: the eta_c analysis presented yesterday by Dima [message #12435 is a reply to message #12428]

Tue, 02 August 2011 16:46

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

first-grade participant

From: *fuw.edu.pl

According to the latest modification done by Lia there are not 2 but 3 mctrackassociator, one before PndSttMvdGemTracking, one after and third after Kalman and all pndtrackid are properly set.

Dima

Report message to a moderator

Re: the eta_c analysis presented yesterday by Dima [message #12436 is a reply to message #12434]

Tue, 02 August 2011 16:49

Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino

first-grade participant

From: *pv.infn.it

Quote:

I added an additional mctrackassociator before the PndSttMvdGemTracking to extrapolate with the MC pdg code.

Lia.

Report message to a moderator

Previous Topic:	Psi3770 study with run981 tpc mode
Next Topic:	Lambda and Lambdabar channel

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Dec 21 17:35:47 CET 2024

Total time taken to generate the page: 0.00689 seconds