hardcoded input parameters in complex mass scheme

[Gauthier]

Dear MG Developers,

When the complex mass scheme is activated, the function "change_electroweak_mode" of madgraph/core/base_obects.py is called and set electroweak parameters to some hardcoded values.

This leads to the very puzzling fact that LO results with complex_mass_scheme set on and off are different. Comparing a LO run without complex mass scheme and a NLO run with it would also lead to wrong k factors.

For consistency, wouldn't it be preferable to fetch electroweak parameters from the model?

Cheers,

Gauthier

[Olivier Mattelaer (olivier-mattelaer)]

Hi Gauthier,

On the principle, I would agree but this is not always the desire way.
If you want to keep the consistency then you would want to keep alpha_ew real and set the other parameter in a consistent way. But if you use complex_mass_scheme, you typically do not care about keeping alpha_ew real, and prefer to have a better control on the mass of the W and Z.

I'm not against any development in that direction but I'm not going to invest time myself in that direction. Now if you provide a patch (or succeed to convince someone to implement such change I would be happy to review those change and include it in the next version of the code)

Cheers,

Olivier

[Gauthier (gauthier.d)]

Hi Olivier,

Thanks for your answer.

I don't have a strong opinion about the fact the EW scheme is changed
from (Gf,mz,α) to (mw,mz,α). I was just thinking that the new input
value of mw, instead of being hard-coded, could be taken from the UFO
(even though it is (Gf,mz,α) based, and therefore obtains it as a
derived quantity).

Cheers,

Gauthier

On 17/08/17 11:28, Olivier Mattelaer wrote:
> Hi Gauthier,
>
> On the principle, I would agree but this is not always the desire way.
> If you want to keep the consistency then you would want to keep alpha_ew real and set the other parameter in a consistent way. But if you use complex_mass_scheme, you typically do not care about keeping alpha_ew real, and prefer to have a better control on the mass of the W and Z.
>
> I'm not against any development in that direction but I'm not going to
> invest time myself in that direction. Now if you provide a patch (or
> succeed to convince someone to implement such change I would be happy to
> review those change and include it in the next version of the code)
>
> Cheers,
>
> Olivier