Dear Stefano de Gironcoli,<br><br><div class="gmail_quote">2011/4/19 Stefano de Gironcoli <span dir="ltr"><<a href="mailto:degironc@sissa.it">degironc@sissa.it</a>></span><br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
have you bothered looking at the output ? has the bfgs procedure<br>
converged ?<br></blockquote><div><br>Yes, I have looked. If you look carefully at one of the previous posts, you'll see<br>that bfgs procedure converged.<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
Notice that in recent versions of vcrelax one additional scf<br>
calculation is performed at the final configuration reinitializing<br>
reciprocal lattice vectors .<br></blockquote><div><br>This has been already explained by Dr. Giannozzi in previous posts too.<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
this has nothing to do with the bfgs procedure. it's a check to<br>
verify what is the impact of the deformation of the cutoff "sphere"<br>
induced by the cell-shape change.<br></blockquote><div><br>My question is exactly about this. More specific: why deformed cutoff<br>
"sphere" gives lower total energy compared to undeformed one?<br>
<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
if the cutoff is large enough the change should be very small. 0.5<br>
mRy looks small to me.<br></blockquote><div><br>Is there any criteria for an energy difference to be called "small"? <br></div></div><br>-- <br>Best regards, Max Popov<br>Ph.D. student<br>Materials center Leoben (MCL), Leoben, Austria.<br>