[Ifeffit] [FEFF8.4] Re(x) is .le. zero in besjn

[Mauro Rovezzi]

Hi Joshua,

joshua jason kas wrote:
> Hi Mauro,
> This is probably a bug in feff. There is probably a workaround.
> Can you send your feff.inp file and I will try to help find the right 
> cards to make your file run. Also I will add this to a list of bugs to 
> look at.

The attached FEFF8 input file works fine with FEFF8.2 and FEFF8.4
compiled on a 64 bit Linux machine and fails on my 32 bit PC. I suppose
that this odd behavior is due to different Fortran compilers used. In
fact, in the first case was used Intel compiler and in the second the
GNU one. I think this kind of problems with Fortran compilers are
frequent so I don't think is a real bug of the FEFF8 code.

M.

-- 
Mauro Rovezzi <rovezzi at esrf.fr>
PhD student - GILDA BM08 c/o ESRF (Grenoble, FR)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: feff.inp
Type: chemical/x-gamess-input
Size: 7780 bytes
Desc: not available
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20071018/881e5d03/attachment.bin>