[Wien] About antiferromagnetism calculation

[Eric Sun]

Dear WIEN2k users,

I'm trying to do some antiferromagnetism calculations.
I read the FAQ in the following page
http://www.wien2k.at/reg_user/faq/afm.html
I realize that an important step in the AFM calculations 
is to manually flip one of the spins of two adjacent
magnetic atoms. My simple question is that
if this step is to avoid the so-called "local minimum"
in the variational procedure. I mean, if we just leave 
the two spins parallel, will the variational procedure
always stop at some ferromagnetic state? Or, 
is it possible that, even if the spins are initially
antiparallel, they may ocassionally drop in 
the ferromagnetic "local minimum" in the variational
procedure?

Thanks for your kindly reply.
Eric
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20050831/c7ca5c41/attachment.html