[CP2K-user] [CP2K:12286] Reproducing results cross-platform

Timo van Opstal tvano... at limebv.nl
Mon Sep 30 07:45:53 UTC 2019


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?

On Sunday, September 29, 2019 at 3:49:50 PM UTC+2, Matthias Krack wrote:
>
> Hi
>
>  
>
> 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.
>
>  
>
> Matthias
>
>  
>
> *Von:* c... at googlegroups.com <javascript:> <c... at googlegroups.com 
> <javascript:>> * Im Auftrag von *Timo van Opstal
> *Gesendet:* Samstag, 28. September 2019 23:38
> *An:* cp2k <c... at googlegroups.com <javascript:>>
> *Betreff:* [CP2K:12286] Reproducing results cross-platform
>
>  
>
> 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!
>
>  
>
> CP2K version: 6.1, SVN revision 18464 
>
> -- 
> You received this message because you are subscribed to the Google Groups 
> "cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to c... at googlegroups.com <javascript:>.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/cp2k/748a2567-9315-4c6d-8b53-4cb2541c2614%40googlegroups.com 
> <https://groups.google.com/d/msgid/cp2k/748a2567-9315-4c6d-8b53-4cb2541c2614%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190930/caaddbea/attachment.htm>


More information about the CP2K-user mailing list