<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto">My question is: how far can you trust the overall accuracy of a numerical dft calculation to that extreme limit? SB<br><br><div dir="ltr"><div><span style="background-color: rgba(255, 255, 255, 0);">___</span></div><div><span style="background-color: rgba(255, 255, 255, 0);">Stefano Baroni, Trieste -- http://stefano.baroni.me</span></div></div><div dir="ltr"><br><blockquote type="cite">On 26 Jan 2021, at 18:24, Zahra Khatibi <za.khatibi@gmail.com> wrote:<br><br></blockquote></div><blockquote type="cite"><div dir="ltr"><div dir="ltr">Dear all, <div><br></div><div><br></div><div>I am currency using ncsf calculation to produce band dispersion. I need to generate high precision eigenvalues, e.g. as small as 0.01 meV. I managed to increase the number of decimal digits shown after the decimal point in the standard output by modifying subroutine <i>print_ks_energies.f90</i> in QE version 6.5 (format 9030 from 8F9.4 to 8F9.5). </div><div>My question is, how far can I go for increasing the number of decimal digits shown in the output, and what parameter dictates the precision of eigenvalues?</div><div><br></div><div><br></div><div>Kind regards,</div><div><span style="color:rgb(136,136,136)">--</span><br></div><div><div><div dir="ltr" data-smartmail="gmail_signature"><div dir="ltr"><div dir="ltr" style="color:rgb(136,136,136)"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div><font size="2">Z. Khatibi</font></div><div><span style="font-family:arial;border-collapse:separate">Postdoctoral fellow</span></div><div><span style="font-family:arial;border-collapse:separate">School of Physics</span></div><div><span style="font-family:arial;border-collapse:separate">Trinity College Dublin</span></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
<span>_______________________________________________</span><br><span>Quantum ESPRESSO is supported by MaX (www.max-centre.eu)</span><br><span>users mailing list users@lists.quantum-espresso.org</span><br><span>https://lists.quantum-espresso.org/mailman/listinfo/users</span></div></blockquote></body></html>