<div dir="ltr">Thank you Matthias! The OSX version still misbehaves after these updates, I've also checked convergence in CUTOFF and REL_CUTOFF, (in the same vein as the tutorial) and it seems that my relative grid error is below 10^-4 in energy for those settings, so that should be OK I think?<br><br>On Sunday, September 29, 2019 at 3:49:50 PM UTC+2, Matthias Krack wrote:<blockquote class="gmail_quote" style="margin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;padding-left: 1ex;">





<div lang="DE" link="blue" vlink="purple">
<div>
<p class="MsoNormal"><span lang="EN-GB">Hi</span></p>
<p class="MsoNormal"><span lang="EN-GB"> </span></p>
<p class="MsoNormal"><span lang="EN-GB">Try again with a smaller EPS_DEFAULT value, e.g. 1.0E-12 instead of 1.0E-7, a larger CUTOFF value, at least 300 (Ry) instead of 200, and a larger REL_CUTOFF value, at least 40 instead
 of 30, to get numerically more meaningful results.</span></p>
<p class="MsoNormal"><span lang="EN-GB"> </span></p>
<p class="MsoNormal"><span lang="EN-GB">Matthias</span></p>
<p class="MsoNormal"><span lang="EN-GB"> </span></p>
<p class="MsoNormal"><b>Von:</b> <a href="javascript:" target="_blank" gdf-obfuscated-mailto="bPNwXZH9CQAJ" rel="nofollow" onmousedown="this.href='javascript:';return true;" onclick="this.href='javascript:';return true;">c...@googlegroups.com</a> <<a href="javascript:" target="_blank" gdf-obfuscated-mailto="bPNwXZH9CQAJ" rel="nofollow" onmousedown="this.href='javascript:';return true;" onclick="this.href='javascript:';return true;">c...@googlegroups.com</a>> <b>
Im Auftrag von </b>Timo van Opstal<br>
<b>Gesendet:</b> Samstag, 28. September 2019 23:38<br>
<b>An:</b> cp2k <<a href="javascript:" target="_blank" gdf-obfuscated-mailto="bPNwXZH9CQAJ" rel="nofollow" onmousedown="this.href='javascript:';return true;" onclick="this.href='javascript:';return true;">c...@googlegroups.com</a>><br>
<b>Betreff:</b> [CP2K:12286] Reproducing results cross-platform</p>
<p class="MsoNormal"> </p>
<div>
<div>
<p class="MsoNormal">I've downloaded precompiled binaries for Windows and Mac OS and run a simple geometry optimisation on a water molecule. However, the output differs significantlly, already the SCF iterations seem to behave better with the Windows binaries.
 I increased the verbose to inspect differences but can't find an apparent reason for the discrepancy between the two runs. Help is much appreciated!</p>
</div>
<div>
<p class="MsoNormal"> </p>
</div>
<div>
<p class="MsoNormal">CP2K version: 6.1, SVN revision 18464 </p>
</div>
</div>
<p class="MsoNormal">-- <br>
You received this message because you are subscribed to the Google Groups "cp2k" group.<br>
To unsubscribe from this group and stop receiving emails from it, send an email to
<a href="javascript:" target="_blank" gdf-obfuscated-mailto="bPNwXZH9CQAJ" rel="nofollow" onmousedown="this.href='javascript:';return true;" onclick="this.href='javascript:';return true;">cp...@googlegroups.<wbr>com</a>.<br>
To view this discussion on the web visit <a href="https://groups.google.com/d/msgid/cp2k/748a2567-9315-4c6d-8b53-4cb2541c2614%40googlegroups.com?utm_medium=email&utm_source=footer" target="_blank" rel="nofollow" onmousedown="this.href='https://groups.google.com/d/msgid/cp2k/748a2567-9315-4c6d-8b53-4cb2541c2614%40googlegroups.com?utm_medium\x3demail\x26utm_source\x3dfooter';return true;" onclick="this.href='https://groups.google.com/d/msgid/cp2k/748a2567-9315-4c6d-8b53-4cb2541c2614%40googlegroups.com?utm_medium\x3demail\x26utm_source\x3dfooter';return true;">
https://groups.google.com/d/<wbr>msgid/cp2k/748a2567-9315-4c6d-<wbr>8b53-4cb2541c2614%<wbr>40googlegroups.com</a>.</p>
</div>
</div>

</blockquote></div>