From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Tue Nov 21 2006 - 15:52:57 CST
I've never done this myself, but I do believe this should work fine
(although you might actually be looking for
impost smthg.so
)
If all you want to do is use some C/C++ routines, you can also use the
extforces command instead of Tclforces, which will dump coordinates to a
file at specified time intervals, run an external program on them, and
read the resulting forces back into NAMD.
Please note, however, that both tclforces and extforces will only run on
the head node, and thus you're likely to suffer greatly if you're
expecting to scale to many processors... if that's the case, only
tweaking the NAMD internals is really going to work well.
Peter
Arturas Ziemys wrote:
> Thanks for answer !
>
> One more general question: Is it possible to import into Tcl Forces some precompiled and wrapped, say C/C++, library to skip slow Tcl numerical calculations ?
>
> And, if YES, is it sufficient to put the like like "import <smthg>" ? There that library should be saved (folder)? Should it be written right after the starting Tcl script:
>
> tclForcesScript {
>
> import <smthg>
>
> ...
>
> }
>
>
> Best
> Arturas
>
>
>
>
>> Hi Arturas,
>> as far as I know this is not possible without editing the NAMD energy
>> function (found in ComputeNonbondedBase2.h), although hopefully future
>> versions will make this easier.
>> Peter
>>
>> Arturas Ziemys wrote:
>>
>>> Hi,
>>>
>>> I am searching ways to introduce the particle which has only repulsive force and the repulsion is applied to some set or types of atoms: the rest atoms are "transparent" to the repulsive forces of that particle.
>>>
>>> Is there any way to implement that without Tcl forces in NAMD ?
>>>
>>> References to some similar implementations would be welcome.
>>>
>>> Best
>>> Arturas
>>>
>>>
>>>
>>>
>
>
>
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:51 CST