[CP2K-user] [CP2K:21613] Re: SCF Energy confusion and problems

Marcella Iannuzzi marci.akira at gmail.com
Fri Jul 4 10:18:54 UTC 2025


Hi ....

I just checked one of the posted output and I see that the calculation has 
not converged.
Then obviously energy and forces are completely wrong. 
Maybe check and understand the information in the output before extracting 
the results.

Regards
Marcella

On Friday, July 4, 2025 at 12:11:13 PM UTC+2 irib... at tcd.ie wrote:

> Hi everyone
>
> I am pretty new to CP2K and I think I am doing something wrong either in 
> the calculations or the analysis... I have a bunch of structures that I 
> have slightly perturbed and calculated energies and forces to train a ML 
> model. However, when I calculate the relative energies between the most 
> stable structure and the rest, the differences are HUGE. I repeated the 
> calculations in ORCA with a similar computational level and the results are 
> really different (and make much more sense to me). 
>
> Could anyone indicate what I am doing wrong? I will add an example here
>
> for this molecule *abdc*, the energy difference between the structure 19 
> (most stable) and the next most stable strute (14) is 372.45kcal/mol while 
> in ORCA is 21.21kcal/mol
> Moreover, the most stable structure in ORCA and CP2K do not even match...
>
> I upload here the input and output files for both calculations in case 
> anyone can help with it.
>
> I am sure I am missing something but I cannot find what it is
>
> Thank you all in advance
>
>

-- 
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 cp2k+unsubscribe at googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/2286ddfe-1a25-4f7f-8fa7-64f57c0f2327n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20250704/4507a58d/attachment.htm>


More information about the CP2K-user mailing list