Frequently Asked Questions
It seems that the ReconstructedParticles branches are empty, but used to be filled.
ReconstructedParticles used to be filled with just smeared info (so analyzers could start developing scripts even if the values in there were not from proper track reconstruction and event building). In the first acadia production we had an unexpected issue filling ReconstructedParticles, but that should be resolved soon. Plan on having access to the actual momentum from track reconstruction in ReconstructedParticles (with exact PID from truth, and energy from momentum and PID).
Should the physics working groups plan to use the calorimeter clusters?
We are still working on full event building that will also take into account energy from clusters, so in acadia the energy is based on the momentum and truth PID.
Which four-vectors should I plan to use?
As for four-vectors, this is somewhat up to you. We are not providing four-vectors in ReconstructedParticles. For some particles they will be four-vectors, though: particles without info from either tracking or calorimetry. Depending on the analysis you are performing, you may want to use momentum direction and energy (say, higher energy central electrons) while for others you may want to use momentum direction and momentum magnitude (say, low energy central electrons).
Why did the files I was using disappear? Will the files disappear?
Files on S3 will not disappear, though RECO files may be replaced by updated reconstruction.
Files on XRootD may disappear when the entire directory is deleted to stay within the required space allocations.
How is GeneratedParticles different from mcparticles?
mcparticles contains the full event generator record, including internal particles that are not in the final state (think: a pi0 with a strong decay into two pions). It is a record that is uses geant4 steps as the basis for its structure, and therefore includes start and end values.
GeneratedParticles contains only the event generator final state particles only (those with HepMC status 1). It is structured in exactly the same way as ReconstructedParticles and allows for comparison of truth with reconstruction.