Created attachment 602 [details] InelasticXS of Gheisha Dear experts, I was comparing the depth-dose with different Geant4 versions by simulating protons interacting with water. I observed the difference between 10.2.p3 and 10.5/10.6. You can see the difference in the attached slides(p.4). To check where the difference came from, I switched off the hadronic processes to do the simulation again, which means the physics processes are only EM. The result(p.3) shows no difference during them. So the above difference came from hadronic processes. To check if the hadronic cross sections are correct, I counted how many times the elastic and inelastic processes happen. The result (p.5) shows that elastic cross sections are consistent during them, but the inelastic cross sections are inconsistent between 10.2.p3 and 10.5/10.6. So the above difference came from the inelastic part. I compared the codes between 10.2.p03 and 10.5/10.6, and found the inelastic cross-section is using the elastic cross-section with Gheisha. You can see the code difference in the slides(p.6). I hope you can take a look at this issue to cross-check if this is a bug. Thanks in advance. Best wishes, Ye Chen
Yes, this is a bug. Fix coming soon. Thanks for finding this.
Hello, the bug is fixed and will be available with the patch to 10.6 release. Thank you very much for the fix. Worse to mention, that it is an old cross section, which is not used in Geant4 10.6 reference Physics Lists. For protons we recommend G4BGGNucleanInelasticXS and G4BGGNucleonElasticXS. VI
