<!DOCTYPE html><html><head> <style type="text/css">body { font-family:'宋体'; font-size:13px}</style> </head> <body><div>Dear Matt,</div><div><br></div><div>Thank you! I did know the importance of the two parameters. So far it seems k point need to be run with atomic guess. I would try modify the multiplicity.</div><div><br></div><div>Yi Wang </div><div><br></div><div><br></div><div>On Tue, 17 May 2016 19:33:15 +0800, Matt W <MattW...@gmail.com> wrote:<br></div><br><blockquote style="margin: 0 0 0.80ex; border-left: #0000FF 2px solid; padding-left: 1ex"><div dir="ltr">It might be worth trying starting from different starting guesses / multiplicities. Possibly they could converge to different solutions.<div><br></div><div>Matt<br><br>On Tuesday, May 17, 2016 at 12:19:37 PM UTC+1, wan...@gmail.com wrote:<blockquote class="gmail_quote" style="margin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;padding-left: 1ex;"><div dir="ltr">Dear Dr. Juerg,<div><br></div><div>Thank you very much.</div><div>The potential seems has been used by the BigDFT team for the delta parameter calculation on deltacodesdft website, with a small error of 2.7 meV/atom compared with Wien2k. According to the results of deltacodesdft, the potential may perform same well in different codes. So I guess maybe the basis is not suitable for bcc-Fe.</div><div>Is there any reference or tutorial you would like to suggest on optimize the basis set?</div><div><br></div><div>Best,</div><div><br></div><div>Yi Wang<br></div></div></blockquote></div></div></blockquote></body></html>