failed to generate Wgamma+2j with SM+AGC model

[Generic User (don't modify these fields)]

I am using Sherpa 1.2.1 in CMS software to generate Wgamma+2j with SM+AGC model. For a non-SM coupling point, I met the following error.

ThresholdMomenta out of bounds !h 18m 33s left ) -> ETA: Sat Jul 03 10:29
Channel_Elements::ThresholdMomenta 2 271.414125 100 49000000 99.9999999985 0
ThresholdMomenta out of bounds !
Channel_Elements::ThresholdMomenta 2 82.419 100 49000000 100 0
ThresholdMomenta out of bounds !7h 24m 46s left ) -> ETA: Sat Jul 03 10:45
Channel_Elements::ThresholdMomenta 1.5 158.629487823 6290.82860176 988232.589046 6290.82860176 0
Channel_Elements::DiceYForward(0.00192900298502,{-8.98847e+307,0,-8.98847e+307,0,0},{-3.12538,3.12538,0.644848}):  Y out of bounds !
  ymin, ymax vs. y : -3.12538 3.12538 vs. -3.12538
Setting y to lower bound  ymin=-3.12538
WARNING in Phase_Space_Handler::OneEvent() :  too many trials for 2_4__G__db__P__nu_tau__tau+__ub
  Efficiency = 0.00140930775169 %.
Sherpa: Matrix_Element_Handler::GenerateUnweightedEvent throws fatal error:
  No weight information.
Exception_Handler::GenerateStackTrace(..): Generating stack trace
{
 0xf7c32637      in 'ATOOLS::Terminate()' (Exception_Handler.C:131)
 0x804b151       in 'main' (Main.C:82)
}
Exception_Handler::Terminate(): Pre-crash status saved to '/home/ncucms/work/CMSSW_3_6_2/src/MySherpa/Wgamma_2j_k15_l00/
test/SHERPATMP/Status__Sat_Jul__3_03-44-39_2010'.

Run.dat and logs are attached.

[Frank Siegert]

It seems you are not isolating the photon from quarks. That's why photon-quark singularities get into your matrix elements which make them very inefficient, and as you have noted also in your other email make the integration times diverge. Try with the following additional line in the selector section:

DeltaR 22 93 0.7 10.

(adjust the 0.7 to whatever kind of photon isolation is loose enough for your analysis)

Does that solve it?