MadGraph5_aMC@NLO

issue with restricted model when mass are opposite to each other

Bug #1813292 reported by Olivier Mattelaer
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
MadGraph5_aMC@NLO
Fix Released
Undecided
Olivier Mattelaer

Bug Description

received by email:

If you set two masses equal and opposite in a model restriction file (i.e. like for neutralinos in the MSSM), then you get a crash at run-time when you check the compatibility of the model parameters with the restriction.
This is because MG5 combines the negative and positive masses that are equal up to a sign into a single parameters, but adding a '-' everywhere in front of that param. But then the check doesn't account for this.
Cheers,

--
Valentin

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) wrote :

I do not reproduce this for the SM:

INFO: Update the dependent parameter of the param_card.dat
WARNING: For model consistency, update mass with id [5] to value 4.7 since it should be equal to the opposite of the parameter with id [6]

will check now for the MSSM

Olivier

Changed in mg5amcnlo:
assignee: Valentin Hirschi (valentin-hirschi) → Olivier Mattelaer (olivier-mattelaer)
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) wrote :

I see the same for the MSSM model...

INFO: Update the dependent parameter of the param_card.dat
WARNING: For model consistency, update mass with id [1000025] to value -181.0882 since it should be equal to the opposite of the parameter with id [1000023]
WARNING: Unknow type of information in the card: vckm,snumix,upmns

The only weird point is that the value written inside the default param_card have those warning showing that it writes the wrong value inside the card. Is that what your problem is?

Cheers,

Olivier

Revision history for this message
Valentin Hirschi (valentin-hirschi) wrote :

Thank you for having looked into this.
Yes, I checked again and, at least in 3.x what I get is a warning, not a crash (sorry for the confusion), so yes this would be the undesired behaviour I am reporting.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) wrote :

Hi,

Here is one solution that fix the issue:
lp:~maddevelopers/mg5amcnlo/2.6.6_bug_1813292
Can you check in your case that this provides the same result as the previous implementation.
(If not I would rather trust this one actually). The only thing that worries me is that it might prevent some optimization (but ok who cares for such extreme case)

Cheers,

Olivier

Changed in mg5amcnlo:
status: New → Fix Committed
Olivier Mattelaer (olivier-mattelaer)
Changed in mg5amcnlo:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.