C. Dai reported about the TRD geometry implementation in Geant4 (see slides).
A. Oliva reported about the SCD working group activities (see slides).
S. Bottai reported about calorimeter saturation studies (see slides).
* J. Casaus asked about the comparison with Fluka and the disagreement with Geant4, and suggested to
simplify the Geant4 geometry to do more direct comparisons. S. Bottai replied that both geometry
and generators contribute to the discrepancies, and that a similar geometry is needed to do precise
comparisons. Anyway the difference is appreciable but not significantly affect the conclusions about
* J. Casaus asked what happens if the shower containment requirement is relaxed. S. Bottai replied that
a good containment is required for good energy measurement, but that the study can be repeated
relaxing the requirement and then correcting for leakage.
* M. Xu said that in Fluka there are some very small energy releases. S. Bottai replied that for Fluka
a real full reconstruction is used, so there are some reconstruction errors, while for Geant4 the MC
truth is used
* M. Xu asked about simulation time at 1 PeV. S. Bottai said that with HerdSoftware the ouptut files
are huge so it was necessary to disable the output for other detectors, which could reduce the CPU
time; with this setup simulating 3-400 protons at 5 PeV with 100 computing nodes took several days.
M. Xu added that 1 PeV proton with Fluka took 1 day for just 1 proton on 1 core.
* A. Surdo said that saturation is usually worse for heavy nuclei, but this does not comes out from
the presentation. S. Bottai confirmed that the same result was obtained with both Geant4 and Fluka,
namely that at constant energy per particle the maximum energy per cube is less for iron than for