Limit in turth information written

[Juanpe]

Hi,

I am using some truth information which is written in the Delphes TTree with the Particle branch but for the information I am using I don't all the particle level information with all hadronisation, with the 40 or 50 first particles would be enough. The problem (or feature) is that since I have to read a huge number of variables per event my analysis takes some time to run. Is it possible to store particle level information only up toa given number of particles? So, limiting the size of the Particle branch not to include all particle generated by hadronisation?

Thanks a lot in advance,
Juanpe.

[Michele Selvaggi]

Hi Juanpe,

the quick and dirty way would be to simply add a break statement in the loop.
If you have a disk space issue, then you have to apply a filter at the time of running delphes.

For instance you could have a look at modules/StatusPidFilter, but you will have to modify it to your own needs.
(as of now, it keeps only particles with status 3, with hard coded pdg codes).

https://cp3.irmp.ucl.ac.be/projects/delphes/browser/trunk/modules/StatusPidFilter.cc

After you have modfied it (and recompiled ..), you can call it in the following way in your data card:

set ExecutionPath {
  MyFilter
  …
}

module StatusPidFilter MyFilter {
add InputArray Delphes/allParticles
add OutputArray filteredParticles
}

…


module TreeWriter TreeWriter {
# add Branch InputArray BranchName BranchClass
  add Branch MyFilter/filteredParticles Particle GenParticle
  ...
}

Hope this helps,
Michele

[Juanpe]

Hi Michele,

thanks a lot! I think it might help indeed. Just a question, since I would like to save the first 50 or so particles, would be enough to exit the while loop in StatusPidFilter.cc::85 after 50 particles have written?

I think so but it is better to check.

Thanks a lot!
Juanpe.

[Juanpe]

Also, another small question, when you say

module StatusPidFilter MyFilter {
add InputArray Delphes/allParticles
add OutputArray filteredParticles
}

you mean set instead of add right? I have never added a new module to a Delphes so I am not sure if that is the correct syntax.

Thanks,
juanpe.

[Michele Selvaggi]

Hi,

yes I meant "set" not "add" (edited).
If you are analysing stdhep files generated with MG5+Pythia6 you can simply set status = 3, which will keep all particles involved in the hard scattering.
If you are using herwig then the conventions for status is a bit different and you might wanna print it out for some events to see what is the best criterium for filtering according to your needs.
This said, probably keeping the first 50 particles should do it for most cases, but better check first.

Michele