[Wien] Small bug in lstart.f

[Peter Blaha]

> >subroutine SOMM:
> >      SUBROUTINE SOMM (DR,DP,DQ,DPAS,DA,M,NP)
> >      ....
> >      DA=0.d0
> >
> >Thus, there MUST not be any difference. (Certainly CTEST should not be
> >initialized by 1.d0, but by 0.d0, but again, it does not matter since it
> >is initialized to zero later on.
> >
>
> well, but the previous line to the 'DA=0.d0' one
> reads:
>
> D1=DA+MM...
>
> and therefore DA is used to define D1 before it is set up to 0.
>
> In fact, I was founding strange core sumes, and setting the ctest1 to
> '0' or '1'
> I got the correct charges.

I'm sorry, I was too quick. You are right this is a "bug" due to an
uninitialized variable.

Fortunately DA (in the input) just regulates the behaviour of the
integrated fuctions at R=0, so the effect of the value of DA should be
very small, but of course it must not be an uninitialized variable.

(Surprisingly enough, non of my compilers ever gave me any problem,
even the old SGI compilers, which were very peculiar about non-initialized
variables....)

                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------