[Wien] force convergence

[Jeff Spirko]

On Thu, Jun 17, 2004 at 12:16:59PM -0700, stargmoon wrote:
>    After  I  did  "run_lapw  -i  50  -fc  0.001", the job stoped after 19
>    iterations,  and  I  got  the  force  as  following (only the last few
>    iterations),

>    :FOR001:           1.ATOM                  0.000                 0.000
>    0.000          0.000
>    :FOR002:     2.ATOM           47.466            0.000            0.000
>    -47.466
>    :FOR003:           3.ATOM                  0.000                 0.000
>    0.000          0.000
>    :FOR001:           1.ATOM                  0.000                 0.000
>    0.000          0.000
>    :FOR002:     2.ATOM          158.872            0.007            0.012
>    -158.872
>    :FOR003:           3.ATOM                  0.000                 0.000
>    0.000          0.000
>    I  don't  understand which number is the real force on atom 2, -47.466
>    or -158.872.

The true force is -158.872.  The forces calculated in all but the
last iteration are estimates that are good enough for the
convergence test.  In the last iteration, the full forces are
calculated, which takes a little longer.

>    I  noticed  that  after  I  did min_lapw to optimize the internal free
>    coordinates,  I did another seperated self-consistent calculation with
>    the optimized coordinates, the resultant forces are different from the
>    ones given by the min_lapw. What's the problem?

Does my explanation above cover this part of your question?

>    Best,
>    Stargmoon

-- 
Jeff Spirko   spirko at lehigh.edu   spirko at yahoo.com   WD3V   |=>

The study of non-linear physics is like the study of non-elephant biology.

All theoretical chemistry is really physics;
and all theoretical chemists know it. -- Richard P. Feynman