[gmx-developers] Re: mdrun

[Erik Lindahl]

Graham Smith wrote:

>Hi -- 
>
>a couple of people have now asked me for a gmx v3 version of the
>modified mdrun I made that can insert a protein into a bilayer - I'm
>going to try to do this at last this weekend, but I'm worried about
>the new version (3.2 or 3.1.5) that you say is about to come out: 
>
>Erik said before that you could probably incorporate the code into
>mdrun; I'm not sure you'd think that if you saw the code - in
>particular it isn't parallel, does an N^2 neighbour search and adds
>about 10 command-line flags (though I'm going to try to fix that at
>least, giving an input param file instead) - and even if you did like
>it I don't know if we've got time to include it before the new
>release? If it might be poss, could I have a deadline for giving you
>the code; if not, then I'd like at least to work with the latest mdrun
>version I can and submit it to you maybe for the release after next; 
>is the current CVS version more or less what the new release is going
>to be? 
>
Hi Graham,

Yes - all the function interfaces of 3.1.x will be there in 3.2 too, so 
it should
only require a recompile. (And possibly changing the include file 
'macros.h'
to 'gmx_macros.h' :-)


Cheers,

Erik