I have a general question.
If I want to perform geometry optimization, then in general I have to use toldfe 10-12, but for ecut, the convergence with accuracy within 1mHa is enough (or such accuracy can not be required)? Which test for ecut should be performed?
Best wishes, Maxim.
choosing ecut for geometry optimization
Moderator: bguster
-
- Posts: 78
- Joined: Wed May 19, 2010 1:17 pm
- Location: Institute of Silicate Chemistry of Russian Academy of Sciences, Saint-Petersburg, Russia
Re: choosing ecut for geometry optimization
I seem to understand it myself. To find out whether a particular etsut is enough for geometry optimization, I have to check convergence of forces in ecut, right?
- Alain_Jacques
- Posts: 279
- Joined: Sat Aug 15, 2009 9:34 pm
- Location: Université catholique de Louvain - Belgium
Re: choosing ecut for geometry optimization
Right ... there is no such a thing as a "good" ecut for any kind of pseudo. You have to check by yourself as for example in http://www.abinit.org/documentation/helpfiles/for-v7.0/tutorial/lesson_paw1.html
kind regards,
Alain
kind regards,
Alain
-
- Posts: 78
- Joined: Wed May 19, 2010 1:17 pm
- Location: Institute of Silicate Chemistry of Russian Academy of Sciences, Saint-Petersburg, Russia
Re: choosing ecut for geometry optimization
Thank you very much, I will surely check it out