[QE-developers] scf treshold parameter in vc-relax (probably a bug)

[Małgorzata Wierzbowska]

Dear QE Team,

I have the feeling that there is a bug in the v.7.2 in the setting SCF
(electronic)
convergence treshold for the vc-relax option.
This user-setting works for the first geometry iteration and falls to the
default value in the next geometry.
It would be good either to keep it as it is set by the user.
Or better, it could be changed down progressively in the following geometry
iterations
in a way given by the user.

Why is it very important?
Immagine a glass of 6-component oxides and rare earth ions and large
cells......
Etotal jumps down by 7 eV in the subsequent geometry and SCF convergence
improves from 0.01 eV to 0.0001 eV but still it is far from default  1^-6
eV.

With best regards,
Gosia


dr hab. Malgorzata Wierzbowska, prof. IHPP PAS

Institute of High Pressure Physics (Unipress)

Polish Academy of Sciences

Sokolowska 29/37, 01-142 Warsaw, Poland

email: malwi45 at gmail.com <malwi at gmail.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20240729/c2bba1d5/attachment.html>