t1 t1~ a returns zero cross section

[Nadja Strobbe]

Dear authors,

We are trying to generate the following process using the latest MG5_aMC@NLO version (2_2_2 or 2_2_3)

import model mssm
p p > t1 t1~ a

The generation of events fails with "Survey returns zero cross section".

When we try the same thing with an older version (I tried 2_1_1), things work without problem, which seems to suggest a bug in the newer versions.

You can find all cards on the following github page:
https://github.com/Saptaparna/LowPtSUSYAnalysis/tree/master/ForNadja

Cheers,
Nadja

[Olivier Mattelaer (olivier-mattelaer)]

Hi,

Thanks for the report. I think that I have identified the line which make the computation to crash.
I’m discussing with the author of those modification to ensure that the fix will work.

In the meantime it is enough to change the line
 1d5 = eamax ! maximum E for the photons
to
 -1 = eamax ! maximum E for the photons

Cheers,

Olivier

On 20 Feb 2015, at 22:03, Nadja Strobbe <email address hidden> wrote:

> Public bug reported:
>
> Dear authors,
>
> We are trying to generate the following process using the latest
> MG5_aMC@NLO version (2_2_2 or 2_2_3)
>
> import model mssm
> p p > t1 t1~ a
>
> The generation of events fails with "Survey returns zero cross section".
>
> When we try the same thing with an older version (I tried 2_1_1), things
> work without problem, which seems to suggest a bug in the newer
> versions.
>
> You can find all cards on the following github page:
> https://github.com/Saptaparna/LowPtSUSYAnalysis/tree/master/ForNadja
>
>
> Cheers,
> Nadja
>
> ** Affects: mg5amcnlo
> Importance: Undecided
> Status: New
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1424077
>
> Title:
> t1 t1~ a returns zero cross section
>
> Status in MadGraph5_aMC@NLO Generator:
> New
>
> Bug description:
> Dear authors,
>
> We are trying to generate the following process using the latest
> MG5_aMC@NLO version (2_2_2 or 2_2_3)
>
> import model mssm
> p p > t1 t1~ a
>
> The generation of events fails with "Survey returns zero cross
> section".
>
> When we try the same thing with an older version (I tried 2_1_1),
> things work without problem, which seems to suggest a bug in the newer
> versions.
>
> You can find all cards on the following github page:
> https://github.com/Saptaparna/LowPtSUSYAnalysis/tree/master/ForNadja
>
>
> Cheers,
> Nadja
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1424077/+subscriptions

[Nadja Strobbe (nadja-strobbe)]

Hi Olivier,

Thanks for the quick response.
We will change that line.

Is it, generally speaking, better to put -1 if we don't want to put any constraint?

Cheers,
Nadja

[Olivier Mattelaer (olivier-mattelaer)]

Hi Nadja,

> Is it, generally speaking, better to put -1 if we don't want to put any
> constraint?

For the moment it should not change anything.
In the long term, this might indeed be use in a different way by our code.
We quite often think to add a pre-survey integration step to allow a better determination of the
bound of integration and starting grid. If we finally implement such method, then it might be more
efficient to put -1 rather than a large value.

Cheers,

Olivier

On 21 Feb 2015, at 18:39, Nadja Strobbe <email address hidden> wrote:

> Hi Olivier,
>
> Thanks for the quick response.
> We will change that line.
>
> Is it, generally speaking, better to put -1 if we don't want to put any
> constraint?
>
> Cheers,
> Nadja
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1424077
>
> Title:
> t1 t1~ a returns zero cross section
>
> Status in MadGraph5_aMC@NLO Generator:
> New
>
> Bug description:
> Dear authors,
>
> We are trying to generate the following process using the latest
> MG5_aMC@NLO version (2_2_2 or 2_2_3)
>
> import model mssm
> p p > t1 t1~ a
>
> The generation of events fails with "Survey returns zero cross
> section".
>
> When we try the same thing with an older version (I tried 2_1_1),
> things work without problem, which seems to suggest a bug in the newer
> versions.
>
> You can find all cards on the following github page:
> https://github.com/Saptaparna/LowPtSUSYAnalysis/tree/master/ForNadja
>
>
> Cheers,
> Nadja
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1424077/+subscriptions