Re: [sdpd] Are the following fair comments about present Rietveld limitations for SDPD?

[Armel Le Bail <armel...@fluo.univ-lemans.fr>]

-------------------------- eGroups Sponsor -------------------------~-~>
Get a NextCard Visa, in 30 seconds! 
1. Fill in the brief application
2. Get rates as low as 2.99% Intro APR with NO annual fee!
http://click.egroups.com/1/9333/14/_/40740/_/969876001/
---------------------------------------------------------------------_->


>"Where Rietveld programs are "behind" compared to Single Crystal 
>Refinement Suites.  (focussing on organic structures)"
>
>1. Lack of routine Structure Visualization

Ortep-3 for windows, Rasmol (see the new RasTop at :
http://www.bernstein-plus-sons.com/software/RasTop_1.3/
able to read CIF files). I would say that some Rietveld
programs may prepare more kinds of output files (.xyz for
instance, this is quite easy), not only SHELX .ins file.

>2. Lack of user-friendly graphical building up of the structure (Q peak 
>   finding - requiring appropriate algorithms for powder diffraction 
>   such as "maximum entropy") 
>
>3. Lack of user-friendly structure refinement by interacting with the 
>   graphical structure on the screen
>
>4. Lack of easy to use powerful point and click restraints and rigid 
>   bodies. (restraints seem mandatory for refining relatively complex 
>   organics from powder data).
>
>5. Lack of easy to use Fourier Map Generation and viewing using appropriate 
>   algorithms such as maximum entropy to clear up the electron density 
>   contour maps.

About point 5., do not forget the powder diffraction handicap : overlapping.
So that Fourier map is good only for quite simple compounds,
otherwise, it is biased, with or without maximum entropy.

>6. Lack of automatic or "point and click" hydrogen placement and refinement 
>   options.

The most important lack is probably the lack of specialized people.
The scientist population in SDPD is not larger than 1000. Given
that only 2% maybe are program developpers, you understand why
powder is "behind". Moreover, 70% of the developers keep their
programs for themselves or try to commercialize them.

Best,

Armel Le Bail
http://www.cristal.org/course/