MadGraph5_aMC@NLO

Can we have npLO and npNLO attributes for ickkw = 0 mode too?

Asked by Yutaro Iiyama

Dear MG5_aMC@NLO developers,

I have a feature request. When we perform UNLOPS merging of LHE from MG5_aMC@NLO in Pythia 8 using the plugin that comes with the package, we need each event in the LHE file to have the npLO and npNLO values set. I'm aware that these attributes are set when we specify "ickkw = 4" in the run card, with mode 4 specifically written for UNLOPS to create what the original paper [1] calls "exclusive events". However, I've found in the past (https://answers.launchpad.net/mg5amcnlo/+question/656617) that ickkw = 4 is problematic for processes like "p p > a j" and "p p > j j" where there has to be a pT cut on the final state of the lowest-multiplicity Born process. So for these processes I have to fall back to ickkw = 0 (i.e. using "inclusive events" in [1]), which means npLO and npNLO flags aren't in the LHE. Currently I'm adding these attributes "by hand" in my code between reading LHE and passing the event block to Pythia, but it would be great if MG5_aMC@NLO already does it for us. Will there be a problem to do so?

Best regards,
Yutaro

[1] https://arxiv.org/abs/1211.7278

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
Rikkert Frederix Edit question
Solved by:
Yutaro Iiyama
Solved:
Last query:
Last reply:
Revision history for this message
Rikkert Frederix (frederix) said : 		#1

Dear Yutaro,

I would rather not. People might have there own LHE readers which might be incompatible with reading these attributes.

Why would it be important to have these npLO and npNLO available in pythia8 for inclusive events? They are non-physical.

Best,
Rikkert

Revision history for this message
Yutaro Iiyama (yiiyama) said : 		#2

Dear Rikkert,

I'm very sorry for asking a question then leaving it for three months..

I understand the concern. I myself had some primitive LHE reader that relies on regular expression and not on proper XML libraries, and that would indeed break with a change like this. We can work around this another way, so I'll drop this feature request.

Just to give the background though, in Pythia 8, when we use the UNLOPS scheme to merge aMC@NLO LHEs of different parton multiplicities, npLO and npNLO values are used to steer the per-event merging parameters.
http://home.thep.lu.se/~torbjorn/doxygen/aMCatNLOHooks_8h_source.html (lines 97-100)