Issue with hepmc from MadGraph + Pythia, qcut dependence

[Joel Jones-Perez]

I am running the MadGraph + Pythia8 + Delphes chain, and I am getting a problem similar to this one: https://cp3.irmp.ucl.ac.be/projects/delphes/ticket/1078

The strange thing is that this problem happens for specific values of qcut. The process I am looking at is p p > nu nu~ + Jets (up to four), with xqcut = 10 GeV. If I set qcut = 15 GeV, there is no problem, and Delphes runs normally. If I change qcut = 20 GeV or 30 GeV, then I get an " ERROR: invalid event format" on the Delphes log.

This also happens if I take the hepmc file from Pythia8 and run Delphes independently.

I am attaching two hepmc files, one with qcut = 15 GeV, which works, and another one that doesn't.

I have reported this at MadGraph (​https://answers.launchpad.net/mg5amcnlo/+question/665561), but have not received any answer so far.

Any ideas? Many thanks in advance.

[Pavel Demin]

I'd say that your problem is very similar to the one reported in

https://cp3.irmp.ucl.ac.be/projects/delphes/ticket/1161
​https://answers.launchpad.net/mg5amcnlo/+question/660911

If MadGraph or Pythia8 generate broken HepMC files, then the problem should be fixed by the MadGraph or Pythia8 developers.

[Pavel Demin]

You don't have to attach the entire hepmc files, the first line that starts with 'E' should be enough.

[Joel Jones-Perez]

Thanks for the info, will have a look.

Coudln't add the files, as they were too large! :-) Here are the first lines:

The one that works (qcut = 15 GeV)

HepMC::Version 2.06.09
HepMC::IO_GenEvent-START_EVENT_LISTING
E 0 -1 -1.0000000000000000e+00 -1.0000000000000000e+00 -1.0000000000000000e+00 0 0 949 1 2 0 113 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1.0808126100290559e+02 1.0808126100290559e+02 -1.0000000000000000e+00

The one that doesn't work (qcut = 30 GeV)

HepMC::Version 2.06.09
HepMC::IO_GenEvent-START_EVENT_LISTING
E 0 -1 -1.0000000000000000e+00 -1.0000000000000000e+00 -1.0000000000000000e+00 0 0 949 1 2 0 115 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1.0808126100290559e+02 1.0808126100290559e+02 -1.0000000000000000e+00

[Pavel Demin]

In the one that works, the number of weights is 113 and there are indeed 113 weights.

In the one that doesn't work, the number of weights is 115 but there are only 114 weights.

So, the error message returned by Delphes is correct and the event format is indeed invalid.

[Pavel Demin]

Since almost all the weights are zeros, a workaround could be manually editing the file and setting the number of weights to a correct value.

[Joel Jones-Perez]

Right. Your link did the trick. I turned off systematics, and everything worked.

Now I have another issue, but it is clearly MG related. Thank you very much!!!