[sdpd] Re: Question about OVERLAP

[Armel Le Bail <alb...@cristal.org>]

Hello,

>When I use the OVERLAP,
>
>I happen to know that sigma(F) is calculated just dividing F with one hundred.
>
>It it reasonable?
>
>I would like to know the principle beneath.   :-)

The principle is simple.

The estimated standard deviations are not at all used in
the process of the Patterson or direct methods.

However, they are used at the beginning of the SHELXS
program, for instance, in order to eliminate some uncertain
reflections.

Sometimes the "|Fobs|" extraction process from powder data
produces crazy estimated standard deviations, so that almost no
"|Fobs|" travels through the SHELXS ESDs test.

Thus, the calculation of the ESDs as "|Fobs|"/100. allows
you to keep the data in the Patterson and Direct methods
process, which, I repeat, does not make use at all of any
ESD. OVERLAP is already a program for elimination of uncertain
reflections.

According to a recent publication in J. Appl. Cryst. by an
Italian team (Altomare, Giaccovazzo, etc), it would be
better to keep all the data for insertion into the direct methods
process : do not eliminate the overlapping equipartitionned reflections,
but repartition randomly the intensity between them. OVERLAP could
be modified easily in order to do that, as an option

Best wishes,

Armel

PS - About programs that pretend to produce correct ESDs for
completely uncertain data (for instance those exactly overlapping
in a powder pattern), I doubt ;-).


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for Your HP, Epson, Canon or Lexmark
Printer at Myinks.com. Free s/h on orders $50 or more to the US & Canada. http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/l.m7sD/LIdGAA/qnsNAA/UIYolB/TM
---------------------------------------------------------------------~->

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/