Statistical Comparison of Particle- and Detector-Level Photon Distributions Obtained with Different PYTHIA Tunes
DOI:
https://doi.org/10.48188/so.7.11Keywords:
PYTHIA, generator tuning, Monte Carlo event generators, detector simulation, particle distributions, statistical analysisAbstract
Aim: To investigate how the choice of tuning in the PYTHIA Monte Carlo simulator influences the properties of final-state particles produced in proton–proton collisions, with particular emphasis on regions of phase space that are not typically constrained during the tuning procedure.
Methods: A total of 6⋅108 proton-proton collisions were generated at center-of-mass energies between 100 GeV and 13.6 TeV. PYTHIA Tunes 05, 14, and 21 were investigated. The resulting event samples were compared using a range of statistical tests applied to particle kinematic distributions. Furthermore, an independent simulation of particle energy deposition in detector components corresponding to real experiments is performed, enabling comparisons of energy distributions in both central and forward detectors.
Results: Significant statistical differences were observed between the distributions obtained with different tunes, both for particle kinematics and for energy deposition in the central and forward detector regions. The analysis identifies the tunings that exhibit the largest deviations from one another. Local deviations of up to 44% in the forward detector energy deposition were observed.
Conclusions: The study demonstrates that tuning choices can lead to measurable differences even outside the regions for which they were optimized. It also estimates the amount of data a real central detector would need to collect to experimentally distinguish between simulated distributions produced by different PYTHIA tunings.
Downloads
Published
License
Copyright (c) 2026 Marko Husar, Nikola Poljak

This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution (CC-BY) 4.0 License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.


