Showering problem for ZZ+2jets NLO with FxFx merging

Dear Lailin,

Your event file has been generated for matching with the 'HERWIG6' parton shower (see the respective parameter in the run_card.dat). Therefore, you'll get unphysical results for showering with pythia8. Indeed, if you use the build-in shower interface from within the MG5_aMC code, the code would have automatically switched to showering with herwig6.

You'll have to regenerate the events using the

PYTHIA8 = parton_shower

option in the run_card.dat.

B.t.w., I believe you need to set JetMatching:nJetMax to 2 instead of 3, because you're only matching up to 2 jets at the Born level.

Best,
Rikkert

Hi Rikkert,

Many thanks for your reply.
I'll change the shoer option to Pythia8 and try it again.

For "nJetMax", maybe a naive question, since I use "p p > z z j j [QCD]" so I still can get 3 jets, right?

Btw, I have another question. I want to do ZZ->llqq decay. At the beginning, I did the decay with madspin
at matrix element level. Then I thought the jets from Z decay might cause matching problem in showering.
So I decided to do the decay in showering process. But I'm not sure if this is the best way. Do you have any
suggestion?

Thanks,
Lailin

Dear Lailin,

> For "nJetMax", maybe a naive question, since I use "p p > z z j j [QCD]" so I still can get 3 jets, right?

Yes, but the 3rd jet is matched using the MC@NLO matching prescription, not the FxFx.

It's okay to used MadSpin. The Pythia merging code knows about jets coming from decays and will exclude them from the merging.

Best,
Rikkert

Hi Rikkert,

Thanks for your clarification and suggestion.

Just another quick question, can I just change the parton_shower option to PYTHIA8 in my LHE file and run Pythia showering,
or do I have to generate a new LHE file?

Thanks,
Lailin

Dear Lailin,

You have to regenerate a new LHE file.

Cheers,

Olivier

Thanks Olivier Mattelaer, that solved my question.

Hi Olivier, Rikkert,

Sorry to come back to this issue again.
I regenerated a LHE file with Pythia8 showering. But I still got the same error as before when I did the showering.

I had the following showering settings for Pythia8:

### FxFx merging
genSeq.Pythia8.Commands += [ 'JetMatching:merge = on',
                             'JetMatching:scheme = 1',
                             'JetMatching:setMad = off',
                             'JetMatching:qCut = 30', ## 2*ptj
                             'JetMatching:coneRadius = 1.0',
                             'JetMatching:etaJetMax = 1000.0',
                             'JetMatching:doFxFx = on',
                             'JetMatching:qCutME = 15.0', ## ptj
                             'JetMatching:nJetMax = 2' ]

genSeq.Pythia8.UserHook = 'JetMatchingMadgraph'
genSeq.Pythia8.FxFxXS = True

The error from Pythia is:

23:49:06 PYTHIA Error in SpaceShower::pT2nearQCDthreshold: stuck in loop
23:50:04 PYTHIA Warning in SpaceShower::pT2nextQCD: weight above unity
23:53:09 PYTHIA Warning in SpaceShower::pT2nextQCD: small daughter PDF
23:54:37 PYTHIA Abort from Pythia::next: reached end of Les Houches Events File
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 INFO Event generation failed - re-trying.
23:54:37 Pythia8 ERROR Exceeded the max number of consecutive event failures.

Do you have any idea about this problem?

My LHE file is at: https://cernbox.cern.ch/index.php/s/lBd9yu8ynvRBDv5

Thanks,
Lailin

Dear Lailin,

I just tried showering your event file and didn't get your error. However, my generated event file was empty: all events were rejected by the FxFx merging. This is obviously not correct. I traced this back to the fact that in your old version (2.3.3 is almost two years old now) there was a problem that MadSpin removes the attributes in the <event> tag, which is then used by pythia8 to determine the number of QCD particles that enter the Matching. In particular, with FxFx merging, the events in the LHE event file look like

"
<event npLO=' -1 ' npNLO=' 0 '>
...
</event>
"

In your event file, they miss the 'npLO' and 'npNLO' information:

"
<event>
...
</event>
"

This makes pythia8 reject all the events. Unfortunately, one cannot reinstate this information without regeneration of the events (sorry about that!).

Even so, all the events were showered (the rejection is done after parton showering), and I did not get your problem. I don't see anything wrong with your setup (apart from the above issue related to the event attributes and MadSpin). If your problem persists with a newer version of MG5_aMC, you might want to contact the pythia8 authors.

best,
Rikkert

Hi Rikkert,

Many thanks for your investigation. I'll try a newer version (2.5.3) and see if the problem
is still there.

Thanks,
Lailin