2015-11-28 4:21 GMT-07:00 Pallavi Banerjee < [email protected]>:
> The potential updates happen one interaction at a time. I do not do them > all together. > > On having looked at the trajectory, I see that the polymer goes from a > coiled up to a very stretched conformation at this step (step 399) very > gradually. > Well, IBI needs a uniform sample of the distribution in each iteration step, meaning if you have multiple phase you need to sample all of them. You could try to increase the number of MD step per iteration. Another idea would be to set cg.inverse.initial_configuration to "laststep" instead of last "maindir" (default), which start each IBI iteration from the finial configuration of the previous step. > > Also, I do not want to stop further iterations because the convergence > values are still far from the convergence limit. They are 57.21 for A-A > interaction, 215.64 for B-B interaction and 69.17 from A-A interaction. I > provide the convergence limit of 0.003. > 0.003 seems a bit ambitious. Also your distributions does not taper off to 1.0, which might lead to other problems, because the U(r_max) is not 0 and hence you cut-off some interaction you might need. And the distributions have very high peaks, which seems different from what I have seen before (Fig 12. of http://pubs.acs.org/doi/abs/10.1021/ma901242h). Christoph > > Is such a situation not a common occurrence ? > > Thanks. > > -Pallavi Banerjee > > > On Friday, November 27, 2015 at 10:57:43 AM UTC+5:30, Pallavi Banerjee > wrote: >> >> Hello all, >> >> I am iterating three non-bonded interactions in my system. I see that >> after having converged to a good extent, the RDFs immediately diverge at >> the next step. Inspite of using a scaling factor of 0.3/0.1/0.001 on the >> potential update, there's no improvement. >> >> Following are the plots for a better understanding. Could someone suggest >> ways to deal with this issue? >> >> Many Thanks! >> >> -Pallavi Banerjee >> >> >> >> >> >> <https://lh3.googleusercontent.com/-D4X_hdiI-Jg/Vlfo9BrmZxI/AAAAAAAAAPc/K5Tc0gp7VTQ/s1600/rdf_A-A.PNG> >> <https://lh3.googleusercontent.com/-cNX1tukV_BY/VlfpEWoMo6I/AAAAAAAAAPk/tGwFguRtYDE/s1600/rdf_B-B.PNG> >> >> >> -- > 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.
