sometime (few percent ?) the process is responsible for the step, but no interaction occurs.
Running tests now.
I have reproduced the problem. For 10 MeV gammas on 100Mo, I get no interaction 7% of the time. As I raise the energy to 15 MeV, this problem goes away, and all appears to fine. The problem is that at these low energies all the gamma-nucleon cross sections on which the Bertini cascade is based are zero. The only non-zero cross sections are the quasi-deuteron ones, where a gamma is absorbed on a nucleon pair. As the projectile energy goes below 15 MeV, the nucleons from the quasi-deuteron absorption are less likely to rise above the nuclear potential. When that happens, the precompound and de-excitation models are called which may or may not produce secondaries. If no secondaries can be produced after several tries, Bertini returns a "no interaction". This is the designed behavior of the model and therefore not a bug. To improve the situation would require a hadronic model of gammma-nuclear reactions in which the nucleus is treated coherently. I believe this would have to be a data-driven model much like HP neutrons.
