[Wien] LDA+U changed between the versions?

[mazin]

I posted a comparison, which awaits moderation, but I am not sure it is 
worth pursuing, because the latest version, 10.1, gives the same result 
as my old calculation in version 07. It seems that the solution is not 
to use version 9, and that's all.

Igor

On 8/3/2010 1:09 PM, Peter Blaha wrote:
> I assumed you are running on different machines ....
>
> So now it is even easier to find out, where the differences come from.
>
> Just run 1 iteration and compare the scf file (and *vorb* files) to see
> where the differences are.
> Do they come from orb, lapw1, lapwdm ???
>
> Without any detailed info, I cannot help.
>
>>> What happens when you copy the clm* dmat* vorb* files from one directory
>>> to the other and continue.
>>
>> As opposed to continuing from the same directory? I did not try, but I
>> am not exactly sure what do you mean. I need to copy all input files as
>> well, and then I do not understand what is the difference compares to
>> just continuing from the same directory.
>>
>>>
>>> Did you just reach another state ?
>>
>> I do. The first iteration after switching the versions shows a very
>> large charge distance, and then it converges - to the same valence
>> state, sure, but the f states are shifted by about 1 eV w.r.t. the
>> d-states (and Fermi).
>>
>>> It could be because of different mixing...
>>
>> I do not see how. I start with the charge densities that in the old
>> version were fully converged, the output was identical to the input.
>>
>> Thanks
>> Igor
>>
>>
>>>
>>>> Thanks for your comments. It is a good idea to download a newer
>>>> version anyway, but unfortunately my problem is of different sort. I
>>>> can converged to the machine accuracy both calculations, WIEN2k_07 and
>>>> WIEN2k_09, and the results are physically different. I have no problem
>>>> with convergence in either case, and can verify manually that both
>>>> calculations are fully converged. The problem is therefore unrelated
>>>> to a possible bug in the mixing scheme.
>>>>
>>>
>>
>

-- 
**********************************************************************
Igor Mazin, NRL, code 6390, 4555 Overlook Ave SW, Washington, DC 20375
Phone: (202) 767-6990; Fax: (202) 404-7546; e-mail mazin at nrl.navy.mil
Home Page http://cst-www.nrl.navy.mil/users/mazin
**********************************************************************