[Wien] DOS spin polarized in WIEN2k-14.2

[Gavin Abo]

I think that I can confirm this problem in WIEN2k 14.2 and probably 
older versions too. It seems to happen when ymax is left as an empty 
field.  I think the problem comes from how the fmax program (fmax.f in 
SRC_trig) tries to find ymax.  On line 21 in fmax.f, it seems like it 
wants to loop from the minimum to maximum x value (i.e., energy) and 
check y values (i.e., DOS values) to see if it is ymax.  However, since 
the ee variable is set to 0, it probably never starts at the minimum x 
value where ymax might exist.  A quick, but possibly sloppy, fix would 
probably be to set the ee variable to a large negative energy value 
(like -999999).

A similar problem might also exist for Pinver.f in SRC_trig, which it 
looks like is used to find ymin when plotting both spins at once with 
neg spin-dn.

On 3/18/2015 5:48 PM, delamora wrote:
> Dear Professor Blaha,
>          I did not answer your mail since you put an earlier date, 20 oct 2014, and I did not see it.
>
>          I do not know what is happening to my WIEN2k, here is the Fe3O4
> DOS-Fe3O4-up   "Do you want to plot both spins at once: YES   with neg spin-dn DOS: YES"
> DOS-Fe3O4-up-dn   "Do you want to plot both spins at once: YES   with neg spin-dn DOS: NO"
>
> DOS-Fe3O4  I edit ":dos2"
> ________________________________________
> De: wien-bounces at zeus.theochem.tuwien.ac.at <wien-bounces at zeus.theochem.tuwien.ac.at> en nombre de Peter Blaha <pblaha at theochem.tuwien.ac.at>
> Enviado: lunes, 20 de octubre de 2014 02:39 a. m.
> Para: A Mailing list for WIEN2k users
> Asunto: Re: [Wien] DOS spin polarized in WIEN2k-14.2
>
> I cannot verify this, neither using w2web or on the command line using
> dosplot2.
>
>
> If you have empty fields for xmin/max and ymin/max is scales automatically.
>
> For dosplot2 there is a    dosplot.ini   file which is used by default
> if it is present (dosplot2 -h)
>
>
> On 10/19/2014 05:00 PM, delamora wrote:
>> Dear WIEN2k users;
>>
>>       I am using the 14.2 version and I am doing spin polarized
>> calculation and when I want to plot
>>
>>
>> 'simple' DOS (without 'both spins at once')
>>
>>       I get a plot with the energy axis limits normal, but the DOS axis
>> is [0:1]
>>
>>
>> both spins DOS (without neg spin-dn DOS)
>>
>>       it is 'the same' that is, the same limits, and both plots, up
>> and dn, are shown
>>
>>
>> both spins DOS with neg spin-dn DOS
>>
>>       I get both axes, energy and DOS with [-1:1], and the the DOS-dn is
>> not shown.
>>
>>
>> Now, if I edit the ':dos2' everything is OK.