[Wien] WIEN2k_04.10

[Shen Li Qiu]

Dear Dr. Blaha,
  At 08:16 AM 10/25/2004 +0200, you wrote:
>I don't think your problems have anything to do with compilation.
>The error indicates a "ghostband", probably comming from a diverged 
>calculation or from not "saving" a calculation,....
>
>
>rm case.scf *.bro*
>x dstart -c
>run_lapw -p

(1) There were case.scf0, case.scf1 and case.scf2 files but there were no
the final case.scf and the *.bro* files because they had not been created
yet when the job was crashed at lapw2.

(2) Following your suggestion, i.e., rm all the *scf* files, x dstart -c
and run_lapw -p, as I expected that it was crashed again at lapw2 in just a
few seconds.

(3) The same job (with the same input STRUCT file, the same initializtion
procedures) ran successfully if the downloaded WIEN2k_04_executables were
used while it was crashed when the WIEN2k_04_executables were produced on
my machine as mentioned in my previous email. 
   With the downloaded WIEN2k_04_executables the calculation was converged
to 10-06 Ry. There was no ghostband at all. So the ghostband is not due to
the input files and parameters but depends on how the executables were
produced. 
That is why I'm wondering some settings on my machine might affect the
WIEN2k_04_executables although there was no error message during the
compiling porcess. Please let me know your suggestions about any possible
reasons that might affect the WIEN2k_04_executables. 

(4) In the worse case that I could not make a workable
WIEN2k_04_executables on my machine would you please compile the
WIEN2k_04.10 code with IDKP=7000 and email the WIEN2k_04_executables to me
in an attached file, because some of my calculations need more that 5000
k-points in the IBZ for making comparison with the results in the literature. 

 Thanks in advance.

 Shen Li Qiu