I have posted a log entry https://logbooks.jlab.org/entry/3469548 regarding simulation of the reaction gp->p pi+ pi- using Geant3 and Geant4. The "measured" resolutions for the pions increase by 50% and tails much more. Is this due to different treatment of the FDCs? See log entry for more details.