The lack of convergence in sampled space should correlate strongly with the error across replicas. Ie this is based on the idea that the dual-topology approach should be using constraints to hold "corresponding" atoms in the same position between the appearing and disappearing matched atoms that have different charges.
The lack of convergence in sampled space should correlate strongly with the error across replicas. Ie this is based on the idea that the dual-topology approach should be using constraints to hold "corresponding" atoms in the same position between the appearing and disappearing matched atoms that have different charges.