MadGraph5_aMC@NLO

Are "special settings for the parton shower" needed?

Asked by Matthew Feickert

Hi. I was recently told by a colleague that the special settings for the parton showers listed on the "Current limitations and common mistakes" section under "Help and FAQs" under https://amcatnlo.web.cern.ch/amcatnlo/

> During the shower phase some parameters are not arbitrary as they have to be identical to those we have assumed in the Monte Carlo subtraction terms. In particular, the following settings MUST NOT BE CHANGED:
>
> * HERWIG6: SOFTME = .FALSE., HARDME = .FALSE., PTMIN = 0.5D0.
> * HERWIGPP (use version 2.4.2 or later and 1.6.1 (or later) for ThePeg): ReconstructionOption=General, InitialInitialBoostOption=LongTransBoost
> * PYTHIA6Q: MSTJ(43) = 3, MSTJ(47) = 0, MSTJ(48) = 0, MSTJ(50) = 2, MSTP(67) = 2, MSTP(68) = 0, PARP(67) = 1D0, PARP(71) = 1D0.
> * PYTHIA6PT: MSTP(67) = 2, MSTP(68) = 0, PARP(67) = 1D0.
> * PYTHIA8 (use version 8.180 or later): SpaceShower:pTmaxMatch = 1, SpaceShower:pTmaxFudge = 1, SpaceShower:MEcorrections = off, TimeShower:pTmaxMatch = 1, TimeShower:pTmaxFudge = 1, TimeShower:MEcorrections = off, TimeShower:globalRecoil = on, TimeShower:limitPTmaxGlobal = on, TimeShower:nMaxGlobalRecoil = 1, TimeShower:globalRecoilMode = 2, TimeShower:nMaxGlobalBranch = 1, TimeShower:weightGluonToQuark=1.

were all necessary.

However, the website https://amcatnlo.web.cern.ch/amcatnlo/ is marked as "no longer up-to-date" under the News view, and hasn't been updated since around 2014 (though hard to say exactly), so given that warning I've always just assumed that it if it isn't on Launchpad then it isn't correct anymore.

There isn't anything listed about special settings for parton showers on the launchpad FAQ page (https://answers.launchpad.net/mg5amcnlo/+faqs), so does that mean these special settings are no longer required, or is the Launchpad FAQ missing information?

For context, this is for/assuming mg5_aMC v3.2.0 for PYTHIA8.

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
Paolo Torrielli Edit question
Last query:
Last reply:
Revision history for this message
Paolo Torrielli (paolotorriell) said : 		#1

Dear Matthew,
these settings are necessary to ensure the correctness of the MC@NLO matching,
so they are still needed.
Best.
Paolo Torrielli

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#2

And just to add to Paolo answer,
the https://amcatnlo.web.cern.ch/amcatnlo/ is still maintained and up-to-date. Only the news section is not used anymore.

Cheers,

Olivier

Revision history for this message
Matthew Feickert (feickert) said : 		#3

> these settings are necessary to ensure the correctness of the MC@NLO matching, so they are still needed.

Thanks for this info. This raises multiple questions then:

>> During the shower phase some parameters are not arbitrary as they have to be identical to those we have assumed in the Monte Carlo subtraction terms. In particular, the following settings MUST NOT BE CHANGED:

Where are these set and when? Are these set in the MG5aMC_PY8_interface? I'm asking because I'm trying to understand if things like this mean that mg5_aMC is actually _requiring_ all external software to be installed and controlled by it for things to run correctly. The `mg5_aMC install` commands seems to give the appearance of helper functions to install software for you if you are missing them. Is this incorrect and mg5_aMC actually needs to be in control over all of its dependencies? If so, then is mg5_configuration.txt not actually meant for users, but is instead supposed to be internal bookkeeping?

> the https://amcatnlo.web.cern.ch/amcatnlo/ is still maintained and up-to-date.

Can you make this **much** more clear in both the website AND Launchpad? The website appears abandoned and Launchpad makes it very unclear what is supposed to be docs.  Also, where can I submit bug reports for the website (here?)? There are numerous links that don't work.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#4

Hi,

You should not confuse the way you install the program from the parameter that you have to set when running the program.
They are zero modification to the pythia8 code applied when installing pythia8, we just sometimes forbids the installer to use the latest pythia8 version if the code is not compatible with MG5aMC (like pythia8.3 is for the moment).
So you can indeed compile pythia8 by yourself and just link it.

MG5aMC_PY8_interface is a pythia8 plugin that is only used for LO computation (madevent) and therefore irrelevant for MC@NLO type of computation. (and irrelevant for this discussion)

If you are running pythia8 from the MG5aMC interface, you will see that the shower_card does not allow you to change all the pythia8 parameter and therefore the comment are automatically satisfied.

If you run pythia8 executable outside of our framework (like many experimental group, some pythia expert,...) then it is your responsibility to configure pythia8 in a way which is compatible with MC@NLO subtraction term (see the MC@NLO paper to understand the physics of such limitation).

> Also, where can I submit bug reports for the website (here?)? There are numerous links that don't work.

Yes you can create a bug report here and I will assign it to the person which has the right to modify it (I can not)

Cheers,

Olivier

Revision history for this message
Paolo Torrielli (paolotorriell) said : 		#5

Dear Matthew,

to complement what Olivier said, those shower settings are explicitly
set in the scripts that steer the showering phase (folder MCatNLO/Scripts),
and are automatically applied when you shower events ‘internally’ e.g.
via the ./bin/shower command).

So, provided you don’t change those Scripts there, the settings will be
correctly read and applied in the shower phase.

If, instead, you use external utilities/interfaces to shower and analyse
events, then these settings must appear in the shower card that you will
create to steer the shower.

Best.
Paolo

> On 21 Oct 2021, at 18:05, Matthew Feickert <email address hidden> wrote:
>
> Question #699173 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/699173
>
> Status: Answered => Open
>
> Matthew Feickert is still having a problem:
>> these settings are necessary to ensure the correctness of the MC@NLO
> matching, so they are still needed.
>
> Thanks for this info. This raises multiple questions then:
>
>>> During the shower phase some parameters are not arbitrary as they
> have to be identical to those we have assumed in the Monte Carlo
> subtraction terms. In particular, the following settings MUST NOT BE
> CHANGED:
>
> Where are these set and when? Are these set in the MG5aMC_PY8_interface?
> I'm asking because I'm trying to understand if things like this mean
> that mg5_aMC is actually _requiring_ all external software to be
> installed and controlled by it for things to run correctly. The `mg5_aMC
> install` commands seems to give the appearance of helper functions to
> install software for you if you are missing them. Is this incorrect and
> mg5_aMC actually needs to be in control over all of its dependencies? If
> so, then is mg5_configuration.txt not actually meant for users, but is
> instead supposed to be internal bookkeeping?
>
>> the https://amcatnlo.web.cern.ch/amcatnlo/ is still maintained and up-
> to-date.
>
> Can you make this **much** more clear in both the website AND Launchpad?
> The website appears abandoned and Launchpad makes it very unclear what
> is supposed to be docs. Also, where can I submit bug reports for the
> website (here?)? There are numerous links that don't work.
>
> --
> You received this question notification because you are subscribed to
> the question.

Can you help with this problem?

Provide an answer of your own, or ask Matthew Feickert for more information if necessary.

To post a message you must log in.