I don't think you'll need so many digits... anyhow if you add a few digits in the output and you tighten the conv_thr to less than 1.d-10 the numbers you'll get should be self-consistent at that level... however, it would be extremely difficult to converge k-point sampling and/or cutoff at that accuracy. not to mention the uncertain error due to the choice of xc.
stefano On 04/13/2011 06:30 AM, Shyam Khambholja wrote: > Thank you Dr. Stefano. > > I need it for calculation of themodynamic properties. > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20110413/ceaf2026/attachment.htm
