-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello Keith--

It seems that XPLOR 98 was an MSI product which was somewhat modified
from X-plor version 3.851. We do not have the source code to XPLOR 98,
and cannot guarantee that XPLOR 98 scripts will work with Xplor-NIH.

> 
> Thank you very much for your response. (John Kuszewski also responded.) 
> I've now encountered the same problem with a single (non-nested) "while" 
> loop. The legacy script (attached),  and a command file that calls the 
> script (attached), work fine using XPLOR 98.1 under UNIX (two minimized 
> structures are produced), but with XPLOR-NIH under Linux, the first 
> structure is produced, and then the while loop error occurs:
> 

what I really need is a complete example, with all input files
included. You should remove any data, however, as this is only a parsing
issue. John and I discussed this briefly, and it seems to be due to
limitations of the @ and @@ control statements, but what exactly is
tripping you up is not yet clear. One or more complete examples will
allow us to reproduce the problem, and likely fix it, or find an easy
workaround.

> 
> Also, is ABNR minimization supported by XPLOR-NIH? (The line "mini abnr 
> nstep=2000 end" was not recognized). Are there alternatives to Powell? I 
> often encounter situations where ABNR works but Powell fails.
> 

It seems that MSI grabbed ABNR from charmm for XPLOR 98. We don't have
it. Implementation of this algorithem within the IVM is on our todo
list. In practise, we've found powell minimization as implemented in the
IVM (accessed using dynamics internal) to be adequate for all purposes
to date.

best regards--
Charles
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>

iD8DBQFHD4bfPK2zrJwS/lYRAhX2AJ9Mcz48Qqe1QHabxZfXa/QIoMhmzwCcDnDd
JvxP6VbfjjqFvvPxx2YQrpw=
=0Q+s
-----END PGP SIGNATURE-----

Reply via email to