Hi Donal,
thanks for pointing this out. I think it is necessary to check the conservation of the extended Hamiltonian for fix rigid/nvt, in a similar way to fix nh. Unfortunately, this calculation has not been done yet.
Steve once suggested us implement a FixRigidNH similar to FixNH, which should allow for implementing a compute_scalar() similar to that in fix_nh.cpp. I don't know if there is any better way to calculate and print out that quantity within the current implementation of fix rigid and fix rigid/nvt.
Steve and other developers may have suggestions for this. I'm willing to add this functionality to fix rigid/nvt once we know the best way to do it.
Cheers,
-Trung
Quoting Donal MacKernan <[email protected]...>: