Event generator and detector simulation softwares

[erkant]

I'm looking for some event generator and detector simulation software. While I was searching, I found a lot of event generator software like PYTHIA, HERWIG, etc. But I was having difficulties in finding a software that will create a detector simulation. If you could tell me which event generator is the best in your opinion, and name some detector simulation software, I would be glad.

 

[Einj]

I'm not an expert, but when I performed a simulation I used GEANT4. It is a software developed by CERN and it allows you to do both event and detector simulations.

 

[jkp]

Another package you might like to look at is MadGraph5. This package allows you to generate Feynman diagrams, calculate cross-sections and generate events. The generated parton level events can be analysed with an addon package called "MadAnalysis" or passed to a showering/hadronization program like Pythia and then passed through a detector simulator such as PGS or Delphes. All these packages can be installed simultaneously with MadGraph5.

One of the nice things about this package is that it can generate events for any theoretical model (SM & any BSM) and there are many model files supplied with the install. New models can easily be defined by the user by hand or you can try the Mathematica package "FeynRules" which can automatically generate MadGraph5 model files once the user defines the correct Lagrangian.

 

[browha]

To answer your question

Basically, no. Not as far as I know. The XML databases used to describe the detectors are massive in size (couple of GB), and because the alignment constantly changes (depending on the experiment, you have thermal effects/etc), you need to be able to time-tag it.

As far as I'm aware, you won't be able to get access to it unless you're part of the experiment. You can try, if you'd like, to get Gauss (for LHCb) or Athena (ATLAS), but I'm 99% sure it will not work.

Event generation, on the other hand, is pretty straight forward. The detector simulation, if you're simulating anything like more than a handful of events, you'll need - at minimum - a cluster

 

[Vanadium 50]

You might look at PGS ("Pretty Good Simulation"). It provides for simplified detector models so you don't have to go through GEANT, which has a very apt name.