trkmcsim duplication per fit hypothesis

[sam-grant]

trkmcsim follows the same structure as the trk branch, plus an extra layer of nesting. So, we have one truth genealogy array per fit hypothesis:

One event:

trk.pdg: [11, -11, 13, -13]
trkmcsim.pdg: [[11, 13, -211, 2212, 11], [11, 13, -211, 2212, 11], [11, 13, -211, 2212, 11], [11, 13, -211, 2212, 11]]

The thing is, the truth genealogy is always the same regardless of the fit hypothesis. Do we need the duplication?

Thanks,
Sam

[AndrewEdmonds11]

Hi Sam,

Thanks for pointing this out. Yes, there is duplication but I think it will be tricky to remove because there could be instances when that isn't the case. For example, if we have two tracks: one gets a successful fit from all fit hypotheses but the second only gets a successful fit from the electron hypothesis, we might have something like:

trk.pdg: [11, 11, -11, 13, -13]
trkmcsim.pdg: [[11, 13, -211, 2212, 11], [11, 22, 2112, cosmic ray], [11, 13, -211, 2212, 11], [11, 13, -211, 2212, 11], [11, 13, -211, 2212, 11]]

which would be easier to handle than:

trk.pdg: [11, 11, -11, 13, -13]
trkmcsim.pdg: [[11, 13, -211, 2212, 11], [11, 22, 2112, cosmic ray]]

so we get an easier-to-use branch, at the cost of duplicated data.

If you can think of a way to have both worlds (i.e. an easy-to-use branch and no duplicationof data), then I'm happy to discuss.

I'll leave this issue open so we have it available for the future

Thanks,
Andy