Re: [sdpd] Seen at Geneva - Call for reviews

[Robin Shirley <R.Shirley...@surrey.ac.uk>]

> Indexing -  Commercial companies (Accelrys, Bruker) make
>                clear efforts (research) in order to break the 
>                TREOR/DICVOL/ITO domination. Increasing
>                computer power allows to propose "brute-force"
>                approaches that needed too high CPU time
>                ten years ago. Real exhaustivity and insensitivity
>                to impurity presence are amost said to be obtained.
>                Will you be able to afford the new software is
>                another story. Many (at least 2 ;-) were asking for the
>                organization of an indexing round robin.

Some further comments on indexing:

1) It confirms that indexing has become a hot area, if commercial
companies are beginning to address it.

2) Accelrys presented 1 program, which like DICVOL uses a form of
successive dichotomy (binary search), but reportedly in a different way.
It remains to be seen whether it will succeed in overcoming the severe
computing-time problems that this method encounters as the search space
expands into low symmetry.  Apparently overnight runs are likely to be 
needed.

3) Bruker presented 2 programs which took a quite different approach by
using Monte Carlo trials followed by rapid refinement and assessment.  
One of them actually made its refinements directly against the observed 
profile rather than peak positions - this is a bold move and it will be 
interesting to see whether it succeeds.

4) But only the commercial packages have been restricted to
ITO/DICVOL/TREOR - other powerful programs have been freely available to
non-profit users for several years (for example KOHL, FJZN & LZON),
notably within the Crysfire suite.

5) A new and much enhanced 2002 version of Crysfire was presented at the
Geneva meeting, and demonstrated in its Software Fayre.  Crysfire 2002
supports 10 indexing programs, including the new MMAP - a graphical
explorer of indexing's solution space.

6) In my opinion, true exhaustivity in general symmetry (including
triclinic) is unlikely to be achieved within the next 10 years, since it
will require an increase in computing speed of a factor of 10000 to
100000 over current GHz processors (which if Moore's law continues to
hold is predicted to take c.20 years or more).  Or need quantum
computers?

7) But I think we can hope for major progress in handling impurity
lines.

8) An indexing round-robin would be very interesting and I'd be happy to
participate.

Robin Shirley

-----------------------------------------

To:            sdpd...@yahoogroups.com
From:          Cristal <alb...@cristal.org>
Date:          Fri, 30 Aug 2002 17:53:16 +0200
Subject:       [sdpd] Seen at Geneva - Call for reviews
Reply-to:      sdpd...@yahoogroups.com

Hello,

For those who were not at the XIX IUCr congress, or at 
EPDIC-8, some SDPD-related comments (personal views) :

Indexing -  Commercial companies (Accelrys, Bruker) make
                clear efforts (research) in order to break the 
                TREOR/DICVOL/ITO domination. Increasing
                computer power allows to propose "brute-force"
                approaches that needed too high CPU time
                ten years ago. Real exhaustivity and insensitivity
                to impurity presence are amost said to be obtained.
                Will you be able to afford the new software is
                another story. Many (at least 2 ;-) were asking for the 
                organization of an indexing round robin.

Direct Space - I was not at EPDIC8 but Yuri Andreev kindly
               sent me his Powerpoint file. He listed 12 programs
               in his review of Direct Space methods, giving details
               about their availability, larger problem solved (etc)
               as furnished by the developer's answers to a
               questionnaire. The 12 program names are :ENDEAVOUR,
               DASH, ESPOIR, TOPAS, POWDERSOLVE, FOX, 
               EAGER, OCTOPUS, PSSP, FOCUS, SAFE, 
               SIMULATED ANNEALING. Wow ! Soon 100 SDPD
               per year, and much more ??

Most speakers at Geneva were not having this sense of the review.
Even with a general title for their conference, they were generally
essentially speaking of their own (or team) results, just as if they
were alone in the world doing that kind of thing.

Other comments :
Being French, I could understand some waitress during
parties speaking about the eager of crystallographers
concerning food : " those people have no education" . The fact
is that sometimes the food could not attain the tables,
particularly during the Israel evening ;-).

Why waiting the IUCr Newsletters for doing your own review
of these congress (we receive these Newsletters too late,
anyway) ? Send here your comments. This may help those
having not the chance to attend. Thanks

Armel



 

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




------------------------ Yahoo! Groups Sponsor ---------------------~-->
4 DVDs Free +s&p Join Now
http://us.click.yahoo.com/pt6YBB/NXiEAA/MVfIAA/UIYolB/TM
---------------------------------------------------------------------~->

 

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