The ABINIT Users Mailing List ( CLOSED )

[Anglade Pierre-Matthieu <anglade@gmail.com>]

Hi,

The message you report means that a minimisation routine has
encountered a situation where it estimates imposible to guess a better
position. Yet fortunately it has find one.

This kind of situation may happen (for instance) in cases where your
plane wave basis set is too small. Or when the SCF loops are not
really converged during a geometry optimisation. For instance, you
shouldn't perform a geometry optimisation without a reasonably  well
converged plane wave energy cutoff (ecut 10 seems a bit small in your
case). You could also benefit from using a smoothing of the plane-wave
cutoff (use ecutsm in you input). Endly you must make sure that your
SCF loops converge correctly between two geometry steps. Usually your
criterion of toldff 5e-5 is sufficient and matches well your tolmxf of
5e-4. However in some difficult SCF convergence case you may get a
toldff < 5e-5 without any real SCF convergence. You can check that by
having a look at VRES2 and  RESIDM in your input. Both of them should
converge steadily at each SCF loop. If they don't you may need to
impose an other SCF convergence criterion. For instance tolvrs 1e-10
or  tolwfr 1e-10 or something of this kind (depending on which one
converges the worst in your case).

Regards

PMA

On Thu, Feb 19, 2009 at 3:28 PM, andika asyuda <venomdi2000@yahoo.com> wrote:
> Good night abinit forum.
>
> I want to ask about following warning I received when running abinit
>
> ---------------------------------------------------------------------------------------------------
>
>  findmin : WARNING -
>   The 2nd degree equation has no positive root (choice=4).
>
>  findmin : COMMENT -
>   Will continue, since the new total energy is lower
>   than the old. Take a larger step in the same direction.
>    line minimization, algorithm   4
>
> ----------------------------------------------------------------------------------------------------
>
> I want to know more about that warning an how to fix my input file.
> I am doing calculation about Fe2O3. Now, I am trying to find the correct
> parameter of ecutoff, so I intend to do several calculation with different
> ecutoff and acell parameter. In my calculation, I used ionmov with value 3.
>
> Please share with me about this warning. Thank you so much. I attached my
> input file to this email.
>
> With regard
> Andika Asyuda
>
>
> ________________________________
> Pemerintahan yang jujur & bersih? Mungkin nggak ya?
> Temukan jawabannya di Yahoo! Answers!



-- 
Pierre-Matthieu Anglade