2015-12-25 4:56 GMT-07:00 Pallavi Banerjee <[email protected]>: > Hello all, > > I have reached a sufficient level of convergence in the rdf profiles. But > the pressure value is way too high. Keeping these converged potentials > fixed, if I start the pressure correction at this step and continue further, > I see that the rdfs are a lot affected even by this linear scheme of update > and the achieved convergence is lost. Does this mean that I need to begin > the iteration procedure from scratch, only this time I include pressure > correction in the iteration process from the very beginning? No, you can just use the last potentials (*.pot.new) and use them as input guesses for a new IBI iteration (by putting them in same directory as settings.xml under the name *.pot.in). VOTCA will use them instead of the simple Boltzmann inversion guess ( U(r) = -k_B*T log g(r) )
> If I were to do > so, what would be a good pattern of pressure correction.? If I have three > rdfs, should the pressure correction be invoked every fourth step , i.e., <0 > 0 0 1> ? It really depends on the system which interactions are strongly correlated and if the IBI update is correlated with the pressure update. However, I would recommend using something like 1 0 0 0, 0 1 0 0, 0 0 1 0 for the three interactions and 0 0 0 1 for the pressure update as a starting and if that doesn't work you could use 1 0 0 0 0 0, 0 0 1 0 0 0, 0 0 0 0 1 0 for the interactions and do three separate pressure update for each interactions using 0 1 0 0 0 0, 0 0 0 1 0 0, 0 0 0 0 0 1. Another idea is to decrease the scaling factor for the pressure correction to not have such a strong perturbation in the rdfs per iteration. Christoph > > Any suggestion is much appreciated,. > > Many thanks. > > -Pallavi Banerjee > > -- > You received this message because you are subscribed to the Google Groups > "votca" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/votca. > For more options, visit https://groups.google.com/d/optout. -- Christoph Junghans Web: http://www.compphys.de -- You received this message because you are subscribed to the Google Groups "votca" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/votca. For more options, visit https://groups.google.com/d/optout.
