Dear all,
This is probably going to sound stupid to most of you, but I am trying to run mkdcd.f from ch2lmp tools folder to convert LAMMPS dump file to CHARMM DCD which is supposed to be a bit different then the one you get when you use dump dcd (I read somwhere the difference is in time units), but I actually don’t know how to run mkdcd.f with the file I want to convert, I don’t know the calling sequence. I am new to this world.
Thanks in advance,
Kemal.
Dear all,
This is probably going to sound stupid to most of you, but I am trying to
run mkdcd.f from ch2lmp tools folder to convert LAMMPS dump file to CHARMM
DCD which is supposed to be a bit different then the one you get when you
use dump dcd (I read somwhere the difference is in time units), but I
actually don't know how to run mkdcd.f with the file I want to convert, I
don't know the calling sequence. I am new to this world.
let me see if i understand you right:
you want to use a software that you cannot figure out how to use because of
some vague supposition that there may be a difference to the
straightforward option without providing any proof that the difference is
relevant. hmm... doesn't make a lot of sense to me. wouldn't it be better
to first run some test whether the difference matters at all??
how to use any of the tools bundled with LAMMPS is usually explained in the
source code or a README file.
most of the stuff in the tools folder is unsupported. the tool in question
is likely to be outdated since according to the svn change log, it has not
been touched since the svn repository was created and the comment in the
header dates it back 10 years. but there *have* been updates and
improvements to the native dump dcd support.
axel.
Well the reason why I am trying to do this, is just because my end goal is to use nMoldyn 3 to get structure dynamic factor for water, but nMoldyn uses as input NetCDF, and it does come with a converter that converts coupled .dcd and .pdb files into MMTK NetCDF. And I have obtained .dcd from lammps and also used lammps2pdb to get pdb, but when I run the converter it literally does nothing, it doesnt start the procedure at all nor does it give me any errors, which I am confused by 
And i have read on it, and it seems like it does charm .dcd so thats why I am asking.
Well the reason why I am trying to do this, is just because my end goal is
to use nMoldyn 3 to get structure dynamic factor for water, but nMoldyn
uses as input NetCDF, and it does come with a converter that converts
coupled .dcd and .pdb files into MMTK NetCDF. And I have obtained .dcd from
lammps and also used lammps2pdb to get pdb, but when I run the converter it
literally does nothing, it doesnt start the procedure at all nor does it
give me any errors, which I am confused by
seems like it does charm .dcd so thats why I am asking.
now *this* is a _completely_ different problem. why don't you say so?
have you contacted the person/people who wrote that converter?
what makes you think that it is the .dcd file and not the .pdb file that is
causing the problem?
have you tried it on some test files generated by some other programs?
are you sure, that you are using the converter correctly?
there are a lot of things that can be wrong.
axel.
btw: after searching for 10 mins using google, i found this:
https://code.google.com/p/dynsf/
https://github.com/pastewka/lammps-netcdf
http://starship.python.net/pipermail/mmtk/2009/001485.html
axel.
The ch2lmp/README and the example files
for mkdcd indicate Paul Crozier helped
develop it. Paul can probably say if
it is still a usable tool with the current
version of LAMMPS.
Steve
Dear Axel,
I have spent a lot of googling on the topic, and I did come across those links too, just at that time I didnt know how to utilize them. But this morning I did successfully (finally,after re-installing ubuntu) install dynsf, and it’s running a calculation right now. When it comes to lammps-netcdf I haven’t been able to install it because of the following step: "
Note that LAMMPS will need to be linked to NetCDF. This will require a
modification of your favorite makefile. Please add
EXTRA_INC += nc-config --cflags
EXTRA_LIB += nc-config --libs
to the respective EXTRA_INC, EXTRA_LIB section of the makefile." I am using Makefile.ubuntu_parallel that I found here: http://katter-world.blogspot.com/2010/05/install-lammps-on-ubuntu.html, and it doesn’t have those sections.
Thanks for all the input and help,
Kemal.
Dear Axel,
I have spent a lot of googling on the topic, and I did come across those
links too, just at that time I didnt know how to utilize them. But this
morning I did successfully (finally,after re-installing ubuntu) install
dynsf, and it's running a calculation right now. When it comes to
lammps-netcdf I haven't been able to install it because of the following
step: "
Note that LAMMPS will need to be linked to NetCDF. This will require a
modification of your favorite makefile. Please addEXTRA_INC += `nc-config --cflags`
EXTRA_LIB += `nc-config --libs`to the respective EXTRA_INC, EXTRA_LIB section of the makefile." I am
using Makefile.ubuntu_parallel that I found here:
http://katter-world.blogspot.com/2010/05/install-lammps-on-ubuntu.html,
and it doesn't have those sections.
both are outdated instructions. the LAMMPS build system has been recently
updated and simplified. you'll be better off reading the documentation that
comes with LAMMPS (it is a mystery to me that some random external
reference seems to have more credit than the original documentation).
anyway, this has *nothing* to do with LAMMPS and all with your skills in
compiling software on linux. that is something you can best learn from some
local people and - if i may add this piece of advice - you *definitely*
should invest some time into that, since that is supposed to be the
foundation of being a computational scientist. if you spend some time on
this now, while you still have time, it will save you a lot of grief later.
axel.
I haven’t used the mkdcd.f tool in a very long time, so I don’t know if it is still compatible with standard LAMMPS dump files. But I imagine that it should be, or would require only minor modification to make it work.
First, you need to compile the fortran source code, then run your executable according to instructions and see what happens.
Paul