The ABINIT Users Mailing List ( CLOSED )

[gonze@pcpm.ucl.ac.be]

Quoting hlil@grenoble.cnrs.fr:

> Please, can sombody confirm me that the following lines included
>  in the loop " do ichan=1,nchan2(iwchan)" wihich is included
> in the loop  "do itemper=1,ntemper":
> 
> gd(ichan)=gd(ichan) + gdos(ichan,iwchan)
> 
> give the phonon dos g(w)

No. gdos is the DOS computed with the histogram method,
but for different channel widths. 
Please, have a look a lines 305 where gg accumulates
the "hits" in each channel, 
and line 475-477 where gdos is normalized
The frequencies belonging to each channels are computed
in lines 292-293. But these are changing for different 
channel widths, of course.
So, you should NOT sum gdos on different iwchan, as you propose.
For each iwchan, you get one histogram DOS.

If this is of no help for you, then you have to wait a developer
can spent times on this, and makes, at last, a user friendly
treatment and output of the phonon DOS. 
However, developers are very busy these days ...

Xavier

-------------------------------------------------
This mail sent through IMP: atlas.pcpm.ucl.ac.be