[gpaw-users] GPAW tutorial

[Daniel Torres]

Guys! the idea of having the cell parameter with just a simple command 
line is awesome. Some comments:
1. Would it be nice to be able to select the number of point the user 
wants in the quadratic regression. I noticed that 5 was used as a default.
2. Would it be nice to be able to see the py file that the command 
generates. In the same way, a general user would like to see the fitted 
data, in order to know by himself if the quadratic correlation is good.
3. Would it be even more nice to have a database with the experimental 
values. I mean, we all use a initial guess from webelements. GPAW could 
automatically use this values as initial guess.
These are just suggestions.
Best regards
Janne Blomqvist wrote:
> Jens Jørgen Mortensen wrote:
>   
>> On Tue, 2010-04-13 at 12:19 +0200, Robert Pennington wrote:
>>     
>>> On 13-Apr-10, at 12:01, gpaw-users-request at listserv.fysik.dtu.dk wrote:
>>>       
>>>> https://wiki.fysik.dtu.dk/gpaw/tutorials/lattice_constants/lattice_constants.html
>>>> does not work with the latest GPAW release? It would be nice to have an
>>>> updated version of the Tutorial section on the GPAW website.
>>>>
>>>>         
>>> The syntax has changed - instead of --kpts=8,8,8 --h=0.18 --fit, try using -k 8,8,8 -g 0.18 -O. I'm not sure if that'll do everything, but it should help somewhat. The help for that command is "gpaw -h".
>>>       
>> Actually, "-k 8,8,8 -g 0.18 -O" works for GPAW version 0.6, but the
>> tutorial uses the new syntax from the latest development version of
>> GPAW.  Yes, it's a bit of a mess!
>>     
>
> Would it be possible to generate, say,
>
> https://wiki.fysik.dtu.dk/gpaw-VERSION/
>
> trees, so that people using the stable versions can easily refer to the 
> documentation of whatever version they're using?
>
> PS: And same for ASE as well, though ASE seems to be quite mature at 
> this stage and thus doesn't have so many incompatible changes.
>
>