sherpa is hosted by Hepforge, IPPP Durham
close Warning: Can't synchronize with repository "(default)" (/hepforge/svn/sherpa does not appear to be a Subversion repository.). Look in the Trac log for more information.
Modify

Opened 14 years ago

Closed 14 years ago

Last modified 10 years ago

#60 closed task (fixed)

Warnings in VBF

Reported by: Frank Siegert Owned by: Stefan Hoeche
Priority: major Milestone:
Component: Unknown Version: 1.2.1
Keywords: Cc: Andrew.Pilkington@cern.ch, kjg@particle.kth.se

Description

In VBF higgs production, roughly 0.1% of events give warnings of two forms (independent of each other):

  • Blob_List::FourMomentumConservation(): (0x1e1a0958) Four Momentum is not conserved. p_{in} = (14000,0,0,0) vs. p_{out} = (14087.1462521,-95.5113582346,-141.371704362,44.202726035).
  • Remnant_Base::TestExtract(db,8297.46641927): Constituent energy out of range E_b = 7000.

We should check where they come from and that they are indeed harmless.

Attachments (0)

Change History (5)

comment:1 Changed 14 years ago by Generic User (don't modify these fields)

Cc: Andrew.Pilkington@cern.ch kjg@particle.kth.se added; Andrew.Pilkington@cern.ch> removed

Hi Frank,

I see these when generating QCD multi-jets (93 93 -> 93 93 93{4}) as well (6 times in 80000 events):

Blob_List::FourMomentumConservation(): (0xaf38cd8) Four Momentum is not conserved.
   p_{in} = (3396.84826289,-4.70875120429e-11,-1.27842377452e-10,432.688891698) vs. p_{out} = (6155.68741318,-2560.52852183,-915.537113763,-32.8957861333).
Multiple_Interactions::CheckBlobList(..): Retry event 3824

Also, one each of these:

Shower::EvolveSinglet(): Reconstruction erm 7s left ) -> ETA: Sun Mar 21 17:39   
or. Reject event.
Hadrons::DiceUncorrelatedKinematics(B+ --> K+ omega(A: Sun Mar 21 17:40  40  
82) pi ): Rejected hadron decay kinematics 10000 times. This indicates a wrong maximum. Will accept kinematics.

Karl-Johan

comment:2 Changed 14 years ago by Stefan Hoeche

Trunk version r14888 fixes most of the problems. Errors still occur for local kin scheme switches with massive initial states if |u-x|<eps. As far as I can see, this is _not_ connected to boost-alignment problems, but rather has to do with the intricate kinematics, leading to xiija=xija*0/0 (numerics in the differences). We may have to come up with a different scheme in these cases or simply discard the points (very few, about 1/3000 events). We could also try and forbid respective clusterings, but a reliable criterion has to be found to identify the problem.

comment:3 Changed 14 years ago by Stefan Hoeche

Resolution: fixed
Status: newclosed

Problems are solved in trunk r14891 and will be included in the next public release. Local kin scheme switches must not occur if ma!=mai as all formulae are derived under the assumption of a conserved spectator mass.

comment:4 Changed 12 years ago by Frank Siegert

Milestone: rel-old

comment:5 Changed 10 years ago by Stefan Hoeche

Milestone: old

Milestone old deleted

Modify Ticket

Change Properties
Action
as closed The owner will remain Stefan Hoeche.

Add Comment


E-mail address and name can be saved in the Preferences.

 
Note: See TracTickets for help on using tickets.