Dear Matt,
It is nice to see a message about SHELX on the BB, compared with the
frequency of messages about certain other programs I have the
impression that no-one is using SHELX these days!
You should be using the larger version SHELXH, available from the SHELX
server. If that is still too small, you need to increase the sizes of
JW and probably IM and recompile; this is explained in a comment at the
beginning of the source (shelxh.f). Recompilation is not very complicated,
e.g. with the Intel compiler under Linux:
ifort shelxh.f shelxlv.f -o shelxh
If you have a very large structure, it might be worth compiling as a
64-bit executable. This would allow larger matrices and reduce rounding
errors. You should also be aware of Kay's multi-CPU version (also
available from the SHELX server) but note that the matrix inversion is
not yet multi-threaded.
Best wishes, George
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Tel. +49-551-39-3021 or -3068
Fax. +49-551-39-22582
On Thu, 28 May 2009, Matthew BOWLER wrote:
> Dear BB,
> I have refined a 1.05A structure with shelxl and would now like to get some
> values on the coordinate error by doing a least squares full matrix cycle,
> unrestrained. I have followed all the procedures that I can find, but I keep
> getting the error message ** ARRAY B TOO SMALL FOR THIS PROBLEM ** even though
> I have added the line BLOC 1 N_1001 > LAST. Is there a way round this? I
> really only want the errors on a ligand so would it help to do the full matrix
> refinement only on a selected set of atoms? Many thanks, yours, Matt.
>
> --
> Matthew Bowler
> Structural Biology Group
> European Synchrotron Radiation Facility
> B.P. 220, 6 rue Jules Horowitz
> F-38043 GRENOBLE CEDEX
> FRANCE
> ===================================================
> Tel: +33 (0) 4.76.88.29.28
> Fax: +33 (0) 4.76.88.29.04
>
> http://www.esrf.fr/UsersAndScience/Experiments/MX/
> ===================================================
>
|