Hi Mohammad, I think, area under the A-B RDF, (specifically \int g(r)4\pi r^2 dr) gives you an estimate of total number of B type neighbours for A type particle, which depends on the A-B interaction. During coarse-graining iterations, A-B pair interaction is modified at each iteration (until convergence), i.e., interaction may not be the same at different iteration steps, and therefore, total number of neighbours and area under g(r) may vary at each iteration step.
Regarding convergence, I also observed, due to statistical nature of short CG-MD simulations at each iteration, CG pair potentials seem to oscillate near convergence after few hundred iterations even though CG RDFs compare well with the target RDFs. So, I suggest you try convergence criteria based on the RDFs instead of potentials. I hope that helps. Best, Sikandar On Sun, Nov 30, 2014 at 10:57 PM, MSW <[email protected]> wrote: > Thanks Christoph for the reply! > > I run my system for 400 iterations and still my system did not converge at > all. The value of the convergence decrease and then increase to some value, > and after that it oscillate around a specific value. I only attached the > first step as an example that the RDF area does not match. > > I thought that RDFs should give similar area! since my system does not > change that much after each iteration, and VOTCA use the same normalization > process. Whats interesting that the shape of the RDF after some iterations > is similar, however, the value is a little bit different. > > You answered my question that the RDFs area should not be equal, however, > I have to discuss this with my PI. > > Thanks again! > > On Sunday, November 30, 2014 6:48:45 PM UTC-6, Christoph Junghans wrote: >> >> 2014-11-30 10:05 GMT-07:00 MSW <[email protected]>: >> > Dear VOTCA users, >> > >> > I am running a system composed of four types of beads in vacuum, and >> after >> > each iteration I look at the RDF and compare between TGT and NEW >> > distribution. I found out that the new RDF does not have the same area >> under >> > the curve, however, they both have the same range for calculation. >> The areas under two arbitrary RDFs are not supposed to be the same, as >> RDF are not area-normalized, but normalized to go to 1 for r to >> infinity. >> >> > >> > The A-A.dist.tgt was calculated using VOTCA (csg_stat) from the >> atomistic >> > data and A-A.dist.new was calculated inside the iteration from CG data. >> This >> > important to look at since convergence in VOTCA is calculated based on >> the >> > difference between the two RDFs. >> > >> > I attached the two RDFs and the current and new potentials (for 1st >> step) >> > with this post. >> Not try to interpret too much in the difference between the RDFs after >> one iteration step. >> If it does not converge after hundreds of iterations, you should be >> worried. >> >> It is also important to note that the convergence is defined very >> differently through out the literature. >> VOTCA supports different measures incl. the l1 and l2 norm. The latter >> was using in the original VOTCA publication. >> >> > >> > I really appreciate your support. >> What is your question actually? >> >> > >> > Thanks >> > >> > Mohammad >> > >> > -- >> > 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 http://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 http://groups.google.com/group/votca. > For more options, visit https://groups.google.com/d/optout. > -- 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 http://groups.google.com/group/votca. For more options, visit https://groups.google.com/d/optout.
