MadGraph5_aMC@NLO

Madevent seems to go into infinite loop for some parameter combinations

Asked by guy pitman

Thank you again for your previous assistance.

I am running a parameter scan on the model in my question 1505897 . For some combinations of parameters the time taken goes from 3 minutes to hours and I have to kill the process. For example running madevent with these parameters

generate_events
set MY0 121.5964
set MXd 20.022757
set WY0 .3
set gsd33 0.00831437
set gsd22 0.6445292
set gsu33 0.0003343066
set xqcut 4.045513

(also with WY0=auto)

It doesn't stop

Is there some way to trace the code to see why this is so? Can you perhaps suggest why this is occurring? Can one put a time limit and force it to exit on these cases?

Many thanks

Guy

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

Some benchmark are more difficult to integrate than other and those will take longer time.
They are no infinite loop at some point MG will automatically give up if this is too difficult for him.

> set xqcut 4.045513

This is extremely low and might be the reason why your run is so slow.

Cheers,

Olivier

> On Oct 30, 2015, at 06:52, guy pitman <email address hidden> wrote:
>
> New question #273399 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/273399
>
> Thank you again for your previous assistance.
>
> I am running a parameter scan on the model in my question 1505897 . For some combinations of parameters the time taken goes from 3 minutes to hours and I have to kill the process. For example running madevent with these parameters
>
> generate_events
> set MY0 121.5964
> set MXd 20.022757
> set WY0 .3
> set gsd33 0.00831437
> set gsd22 0.6445292
> set gsu33 0.0003343066
> set xqcut 4.045513
>
> (also with WY0=auto)
>
> It doesn't stop
>
> Is there some way to trace the code to see why this is so? Can you perhaps suggest why this is occurring? Can one put a time limit and force it to exit on these cases?
>
> Many thanks
>
> Guy
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
guy pitman (guy-pitman) said : 		#2

Thank you!

I did this because of
  xqcut: minimum jet measure (pT/kT) for QCD partons, if xqcut=0 use cone jet matching, if xqcut> 0 use kt jet matching. This value should be related to the hard scale (e.g. mass of produced particle, HT cut, or similar) in the process, and set to ~(1/6-1/3 x hard scale). Please check that the differential jet rate plots (which are automatically generated if you have MadAnalysis and Root properly installed on your system) are smooth, and check that the cross section does not vary significantly when the xqcut is varied up and down.

ie xqcut was: 1/5 of Mxd.

A few months ago I ran the same parameter scan on MG5_aMC_v2_2_3 and it hung on a similar set of parameters, only that time it was because I had xqcut=15? I found that by making xqcut the lesser of 15 and MXd/5 it ran through? Has something changed , or am I possibly missing something?

Thanks again

Regards
Guy

Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#3

If it provides correct results then that’s ok.

Cheers,

Olivier
> On Oct 31, 2015, at 07:02, guy pitman <email address hidden> wrote:
>
> Question #273399 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/273399
>
> guy pitman posted a new comment:
> Thank you!
>
> I did this because of
> xqcut: minimum jet measure (pT/kT) for QCD partons, if xqcut=0 use cone jet matching, if xqcut> 0 use kt jet matching. This value should be related to the hard scale (e.g. mass of produced particle, HT cut, or similar) in the process, and set to ~(1/6-1/3 x hard scale). Please check that the differential jet rate plots (which are automatically generated if you have MadAnalysis and Root properly installed on your system) are smooth, and check that the cross section does not vary significantly when the xqcut is varied up and down.
>
> ie xqcut was: 1/5 of Mxd.
>
> A few months ago I ran the same parameter scan on MG5_aMC_v2_2_3 and it
> hung on a similar set of parameters, only that time it was because I had
> xqcut=15? I found that by making xqcut the lesser of 15 and MXd/5 it
> ran through? Has something changed , or am I possibly missing
> something?
>
> Thanks again
>
> Regards
> Guy
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
guy pitman (guy-pitman) said : 		#4

Thanks Olivier Mattelaer, that solved my question.