Hi, Xianghuai Wang,
Maybe you have misunderstood the meaning of integration. Many integals in
ADF are evaluated by numerical integration, and it requires only one input
parameter which determines the precision of numerical integrals and derives
from that the number of integration points. For single point and geometry
optimization, the default integration is 4.0.
The subkey converge define the convergence thresholds for geometry
optimization. The relationship between them is for a high converge creteria
we also need a high integration. In default, the program automatically
select integration in each step according to the given number.
Please check the adf manual for details.
Hongjun Fan
fan_at_chemie.uni-siegen.de
----- Original Message -----
From: "schwarz" <schwarz_at_chemie.uni-siegen.de>
To: "Fan Hongjun" <fan_at_chemie.uni-siegen.de>
Sent: Tuesday, August 07, 2001 8:05 AM
Subject: [Fwd: ADF-LIST: Convergence and Integration]
>
>
> xianghuai wang wrote:
> >
> > Dear all,
> > I am a new user in adf. I found in adf input file,
> > geometry
> > convergence 0.1
> > end
> >
> > It will automatically set Grad=0.1 which the manual doesn't mention.
> > My question is then how to set the integration? In Cartesian, there will
be
> > 3 number of convergence: E, Grad, Rad.
> > I should set it to
> > Integration 4.5 2.5 3.5
> > or just set it as
> > Integration 2.5
> > ?
> >
> > Thank you very much!
> >
> > _________________________________________________________________
> > 您可以在 MSN Hotmail 站点 http://www.hotmail.com/cn 免费收发电子邮件
>
> --
> *****************************************************************
> Prof. Dr. W.H.Eugen Schwarz
> Labor f黵 Physikalische und Theoretische Chemie
> AG Theoretische Chemie Tel: (+49)-271-740-4338/9
> Universitaet Siegen FAX: (+49)-271-740-2851
> D-57068 SIEGEN e-mail: schwarz_at_chemie.uni-siegen.de
> Germany http://www.tc.chemie.uni-siegen.de/
> ******************************************************************
>
Received on 2001-08-07 16:02:04
This archive was generated by hypermail 2.2.0 : 2006-11-02 07:00:02 CET