Missing discretesampler.mod

Asked by Li on 2020-07-04

Hi,
    I follow the manual step in https://answers.launchpad.net/mg5amcnlo/+faq/2243 , and in the step 4) , after using "make", I got a error,

gfortran -O -w -fbounds-check -fPIC -ffixed-line-length-132 -w -c driver.f -I../../Source/
driver.f:206:9:

       use DiscreteSampler
         1
Fatal Error: Can't open module file ‘discretesampler.mod’ for reading at (1): No such file or directory
compilation terminated.
makefile:60: recipe for target 'driver.o' failed
make: *** [driver.o] Error 1

    Do you know how to solve this issue?
    Thank you!

Best,
Li

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Li
Solved:
2020-07-09
Last query:
2020-07-09
Last reply:
2020-07-08

Why not using 2.7.3 and the new features for that?
Everything should be automatic

Cheers,

Olivier

Li (huangli-itp) said : #2

Hi Olivier,
    Is there a way to generate the interference with the automatic way? For example if I want to generate W{0}W{0}, W{T}W{T} and their interference.

Best,
Li

No this is not possible.

To compute such interference directly you would need
1) to let the W to decay (onshell propagator does not have interference)
2) have a dedicated model with a second copy of the W particles and different coupling order
3) use propagators.py to define a specific polarization to each particles
4) use the intereference syntax (not compatible with decay-chain syntax)
5) have some cut on the final states (otherwise the interference will be zero)
6) implement cut to garantee that the W are close to be onshell (otherwise the full computation does not make sense)
7) hope that the phase-space integration will work (likely it will not)

This has never been tried.

Cheers,

Olivier

PS: You can obviously get it by a substraction of contribution but you are likely to suffer issue with numerical precision

> On 6 Jul 2020, at 22:05, Li <email address hidden> wrote:
>
> Question #691678 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691678
>
> Li posted a new comment:
> Hi Olivier,
> Is there a way to generate the interference with the automatic way? For example if I want to generate W{0}W{0}, W{T}W{T} and their interference.
>
> Best,
> Li
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Li (huangli-itp) said : #4

Hi Olivier,
    Thank you for the detail reply.

    Can we use the narrow width cut "bwcutoff" for 6) and make sure W close to onshell? If not, can we use a cut on "pt_{lvl}", similar to the ptll ( but only runs for same flavor lepton and neutrino ) to make sure that?
    I don't know how MG5 deal with the weight, do you mean the negative weight may lead a failure integration?

    For the 1)~5), thank you for the notice, and I will try them.

    Also, I am more interested in the differential cross section for the interference part but not the total cross section. So I just want to know more about how MG5 generate these events.

Best,
Li

Li (huangli-itp) said : #5

BTW, I accidently find if I run madevent once, the "make" works.

Best,
Li

Hi,

> Can we use the narrow width cut "bwcutoff" for 6)

Since decay-chain syntax are not compatible with interference syntax, you can not use that trick.

> Also, I am more interested in the differential cross section for the
> interference part but not the total cross section. So I just want to
> know more about how MG5 generate these events.

I guess this is the paper that you want to read: hep-ph/0208156 <https://arxiv.org/abs/hep-ph/0208156>
This does not cover interference, but we use that exact method for interference as well
(this is why interference can badly behaves)

Cheers,

Olivier

> On 9 Jul 2020, at 00:31, Li <email address hidden> wrote:
>
> Question #691678 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691678
>
> Li posted a new comment:
> Hi Olivier,
> Thank you for the detail reply.
>
> Can we use the narrow width cut "bwcutoff" for 6) and make sure W close to onshell? If not, can we use a cut on "pt_{lvl}", similar to the ptll ( but only runs for same flavor lepton and neutrino ) to make sure that?
> I don't know how MG5 deal with the weight, do you mean the negative weight may lead a failure integration?
>
> For the 1)~5), thank you for the notice, and I will try them.
>
> Also, I am more interested in the differential cross section for the
> interference part but not the total cross section. So I just want to
> know more about how MG5 generate these events.
>
> Best,
> Li
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Li (huangli-itp) said : #7

Hi,
    Thank you very much!

Best,
Li