Weights with gridpack

Asked by Amin Aboubrahim on 2018-01-23

Hi Olivier,

I am generating events with gridpack while setting event_norm to average in the run_card. For some reason, the lhe file I get has weight as cross-section per number of events, i.e. the sum of weights is the cross-section.
When I do a normal run (with no gridpack) I get the correct normalization.

1) Does gridpack have a pre-set normalization method?
2) After a normal run, the lhe file has the weights as the average cross-section as I asked for but when passed to Pythia and then Delphes, the root file I get has weights whose sum is the cross-section. Are you aware of this? Does matching and merging with Pythia change the event normalization?

Thank you,
Amin

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
2018-01-24
Last query:
2018-01-24
Last reply:
2018-01-24

Hi Amin,

I have done the following:

generate p p > t t~
output
launch
set gridpack T
set event_norm average

the associate cross-section is
 Cross-section : 504.1 +- 2.832 pb

Then I unpack the gridpack and run it via
./run.sh 123 123

then I see the following init block:
<init>
2212 2212 6.500000e+03 6.500000e+03 0 0 247000 247000 -4 1
5.082300e+02 3.257243e+00 5.082300e+02 1
<generator name='MadGraph5_aMC@NLO' version='2.6.1'>please cite 1405.0301 </generator>
</init>

and the following weight for the event:
+5.0823000e+02

So everything seems ok here.

My guess is that you request PY6 to be on when generating the gridpack, since our interface withPY6 does not support "average" that parameter is overwritten to "sum" automatically (you should have a warning about it).
I guess that this also answers your second question actually.

Cheers,

Olivier

Amin Aboubrahim (amin83) said : #2

Dear Olivier,

Thanks for the clarification. You have guessed right, I am switching PY8 on.

Best,
Amin

Amin Aboubrahim (amin83) said : #3

Thanks Olivier Mattelaer, that solved my question.