Project

General

Profile

Actions

Development #320

closed

CbmTrackMatchNew problem with Mvd only tracks.

Added by Philipp Sitzmann almost 7 years ago. Updated over 6 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Target version:
Start date:
07/06/2015
Due date:
% Done:

100%

Estimated time:

Description

There seems to be a problem with mvd only tracks and track matching.

If I use the latest Trunk and loop over all tracks and get the match there are some tracks with a trackMatch without a link.
As I found out the CbmStsTrack has 0 StsHits but 3 or 4 MvdHits:

This is some debug info I included in the Open charm track selection because i had some strange behaviour.

[INFO ] match without links found
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 3
[INFO ] CbmD0TrackSelection: StsTrack: 909 of 977
[INFO ] match without links found
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 3
[INFO ] CbmD0TrackSelection: StsTrack: 910 of 977
[INFO ] match without links found
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 3
[INFO ] CbmD0TrackSelection: StsTrack: 911 of 977
[INFO ] CbmD0TrackSelection: StsTrack: 912 of 977
[INFO ] match without links found
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 4

in the output i print

...
...

stsTrack = (CbmStsTrack*) fStsTrackArray->At(itr);
mcTrackMatch = (CbmTrackMatchNew*) fStsTrackMatches->At(itr);

if(mcTrackMatch->GetNofLinks() > 0)
mcTrackIndex = mcTrackMatch->GetMatchedLink().GetIndex();
else {
LOG(INFO) << "match without links found" << FairLogger::endl;
LOG(INFO) << "Nr of Hits total: " << stsTrack->GetNofHits() << " Nr of mvd hits: " << stsTrack->GetNofMvdHits() << FairL$
continue;
}

...
...

without the if else the programm crashes at the first match without sts hits.

regards
Philipp


Related issues

Related to CbmRoot - Development #329: Revise StsTrack classClosedJulian Book09/10/2015

Actions
Actions #1

Updated by Philipp Sitzmann almost 7 years ago

Had to include CbmMatchRecoToMC::SetIncludeMvdHitsInStsTrack(kTRUE);
this fixed the crash with empty links in CbmMatches.

still the open question is that there are more mvd hits then total hits:

[INFO ] CbmD0TrackSelection: StsTrack: 904 of 978
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 3
qp: -8.19764
[INFO ] CbmD0TrackSelection: StsTrack: 905 of 978
[INFO ] Nr of Hits total: 2 Nr of mvd hits: 3
qp: -nan
[INFO ] CbmD0TrackSelection: StsTrack: 906 of 978
[INFO ] Nr of Hits total: 1 Nr of mvd hits: 3
qp: -10.2526
[INFO ] CbmD0TrackSelection: StsTrack: 907 of 978
[INFO ] Nr of Hits total: 1 Nr of mvd hits: 3
qp: -nan
[INFO ] CbmD0TrackSelection: StsTrack: 908 of 978
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 3
qp: -9.51683
[INFO ] CbmD0TrackSelection: StsTrack: 909 of 978
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 3
qp: -1.80986
[INFO ] CbmD0TrackSelection: StsTrack: 910 of 978
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 3
qp: 0.978589
[INFO ] CbmD0TrackSelection: StsTrack: 911 of 978
[INFO ] Nr of Hits total: 1 Nr of mvd hits: 3
qp: -4.7935
[INFO ] CbmD0TrackSelection: StsTrack: 912 of 978
[INFO ] Nr of Hits total: 0 Nr of mvd hits: 4

I am not really sure if this is a correct behaviour or an bug.

Actions #2

Updated by Volker Friese almost 7 years ago

  • Tracker changed from Bug to Development
  • Status changed from New to Feedback
  • Assignee set to Philipp Sitzmann
  • Target version set to NOV15

The behaviour you quote is the expected one. It is a result of the recent inclusion of Mvd Hits into the CbmStsTrack class.

Now, the StsTrack contains two vectors of hit indizes: one for STS (from the mother class CbmTrack), one for the Mvd. The accessor GetNofHits() (also from the mother class) gives the size of the inherited vector, i.e. the number of STS hits, and not the sum of STS and MVD hits.

I agree this is misleading. What could be done is:

  • Add a method GetNofStsHits(), to CbmStsTrack, returning the size of the STS hit vector;
  • Overload the method GetNofHits() from CbmTrack in CbmStsTrack (would have to be made virtual then), returning the sum of the numbers of STS and MVD hits.

Would you agree to that?

Actions #3

Updated by Volker Friese almost 7 years ago

  • Status changed from Feedback to Resolved
  • Assignee changed from Philipp Sitzmann to Volker Friese

Sorry, I just saw that this issue was followed up by #329.

Actions #4

Updated by Volker Friese almost 7 years ago

  • % Done changed from 0 to 100
Actions #5

Updated by Volker Friese over 6 years ago

  • Status changed from Resolved to Closed
Actions

Also available in: Atom PDF