Can I fixed luminosity ? Can I plot other diagram?

Hi,

> Can I fixed Luminosity?

Yes and No, you can fix the total number of events, which is the same (up to statistical fluctuation) than fixing the Luminosity,

Let me explain this table (the main purpose is debugging in fact):

> Graph Cross-Section Error Events (K) Unwgt Luminosity
> G1 6.284e-09 7.26e-11 7001.0 360.0 5.73e+10

1) Graph: This correspond to the channel integration based on some of the diagram.
                  See the madevent paper (http://inspirehep.net/record/593054) for more information and the following thread for technical detail:
                  https://answers.launchpad.net/madgraph5/+question/187387
2) Cross-section: cross-section associate to the graph.
                 This cross-section is everything but physical (only the sum of all the graphs is a physical cross-section)
3) Error: numerical error associate to the channel
4) Events: Number of phase-space point passing the cuts. (This is just a technical quantity)
5) Unwgt: Number of phase-space after a (local) unweighting
6) Luminosity: Unwgt/cross. Indicates that the number of unwgt event correspond to a given luminosity.

After a first iteration (that we call survey) we refine all the graph that are needed in order to reach the total number of unwgt event.
The graph with more events than required, will not be refine, and those one might a huge Luminosity. All those who have a small luminosity
after the first iteration will be increase to reach the target luminosity/number of events.

So in the ideal case, the luminosity should be close to a constant at the end of the run (which means that the code is 100% efficient)

> Can I plot other diagram?
> I trying to launch process, and than make cross section - parametter diagram.
> Can I plot Unwgt or Luminosity - parametter diagram?

This I'm not sure to understand your question.
All those numbers are rather technical and are not physical, (of course they are related to physical quantity, so they are useful for us to understand what happens in case of trouble with the phase-space integration).
For example, All those number can be changed by a factor 10 or more --including the cross-section associated to a given graph--
from a version of MG5 to the next one. This happens when we change the optimization of the code. Even the number of graph can changed from one version to another.

Cheers,

Olivier

On Nov 19, 2013, at 3:31 PM, Chia-Feng Chang <email address hidden> wrote:

> New question #239483 on MadGraph5:
> https://answers.launchpad.net/madgraph5/+question/239483
>
> Hi, MadTeam
>
> Can I fixed Luminosity?
> For example : generate p p > h > z z ,
> that'll show different luminosity in process u u~ > h > zz , d d~ > h > zz , c c~ > h > zz ..
>
> What's Events maining ?
> For example : in Process results
> Graph Cross-Section Error Events (K) Unwgt Luminosity
> G1 6.284e-09 7.26e-11 7001.0 360.0 5.73e+10
> that show "Events(K)" always different with events in run_card.dat.
> (may events is physical in run_card.dat?)
>
> Can I plot other diagram?
> I trying to launch process, and than make cross section - parametter diagram.
> Can I plot Unwgt or Luminosity - parametter diagram?
>
> Thank you!
> Chia-Feng Chang
>
>
> --
> You received this question notification because you are a member of
> MadTeam, which is an answer contact for MadGraph5.

Hi , Olivier Mattelaer ,

Thank for your Answer again!

Chia-Feng Chang