Page 2 of 2

Re: NaNf in ndb.RIM_W causing NaN's in QP energies

Posted: Fri Apr 28, 2023 9:05 am
by Daniele Varsano
Dear Franz,
I can's spot anything wrong in your report setup.
Looking at the report of PPA posted some days ago, I see you are using XTermKind= "BG" .
Can you repeat the calculation and see the behaviour of eps^-1 setting XTermKind= "none".
in general, I suggest to not use the Xterm, it is not robust and it is memory intensive, so at the end of the story you not gain much.
I do not know if it is causing problem and giving you the unphysical behaviour.

Best,
Daniele

Re: NaNf in ndb.RIM_W causing NaN's in QP energies

Posted: Wed May 03, 2023 10:11 am
by Franz Fischer
Hi Daniele,

I repeated the calculation without using the BG Xterminator and it seems that my problem is resolved, as the inverse dielectric function looks as I expected (see attached figure) and the RIM_W method gives numerical values that also seem reasonable.

Thanks a lot for your help.

Best,
Franz

Re: NaNf in ndb.RIM_W causing NaN's in QP energies

Posted: Wed May 03, 2023 1:33 pm
by Daniele Varsano
Dear Franz,

great it is solved and thanks for reporting, we will have a look to understand what's wrong with the Xterminator.

Best,
Daniele