VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Fri Aug 13 2010 - 15:18:58 CDT

Hi Axel,
  I agree it is time to upgrade the VRPN library that
I compile VMD against, in order to make it easier on users
of new devices like the Novint Falcon. The main thing I will
need to do in order to accomplish this is to get all of my group's
VRPN servers recompiled using the latest VRPN and verify that
our Sensable Phantoms, Magellans, and other devices all still
run correctly. Once I've done that, I will be in a position to
upgrade the VRPN client library that I compile all of the
pre-built VMD binaries against, and we should be in good shape.

Russell: Is the VRPN network protocol still evolving regularly,
or can we expect that by linking VMD against one of the latest
revs, that it should last us for a year or two for the life of the
next major VMD release?

Cheers,
  John

On Fri, Aug 13, 2010 at 03:30:28PM -0400, Axel Kohlmeyer wrote:
> all,
>
> some additional info.
>
> i have since received one report that there _can_ be a communication
> problem between the precompiled VMD versions and a newer vrpn_server
> that was cured by compiling VMD from source. from the details i was given
> it would be very platform specific, and likely a 64 bit issue.
> i suppose some people get lucky, some don't.
>
> i've decided to make an effort to make things easier for people
> that struggle with compilation of libnifalcon and the falcon enabled
> vrpn_server. i've installed static library packages on my build machines,
> adjusted configurations and produced a set of packages with vrpn
> binaries that have only minimal dependencies on external libraries.
> i have a 32bit machine running fedora 10 and a 64bit machine running
> fedora 12 that i am testing on. you can find those pacakges here:
>
> http://klein-group.icms.temple.edu/akohlmey/vrpn-icms/
>
> there are two variants of the vrpn packages. one that is build
> off my local svn that was based on the latest release version
> of VRPN (7.26.0) and another build that is compiled from the
> latest vrpn public git repository plus my modifications (7.26.2).
> in the future i will stay in sync with that repository as close as
> possible and russ will hopefully at some point merge in my
> changes into his tree, so that the code will be more widely
> accessible. for more details have a look at the page where
> i collect all info and examples on my branch of the code.
>
> http://sites.google.com/site/akohlmey/software/vrpn-icms
>
> john, what do you propose would be the best way to get
> some proper VMD builds with a recent VRPN client included?
> i can created distributions from what i am using, but there
> are lots of customizations in my tree and the binaries use
> system libraries for tcl/tk/fltk/python and more, so they are
> likely not portable beyond a replica of my installation.
>
>
> cheers,
> axel.
>
>
> On Fri, Aug 13, 2010 at 1:55 PM, John Stone <johns_at_ks.uiuc.edu> wrote:
> > Shawn,
> >  If you're using the most recent version of VRPN, then you'll likely
> > have to compile your own VMD from source code, as we use a slightly older
> > VRPN library in our VMD builds, to go along with drivers for the
> > Sensable Phantom devices.  None of the error messages you were getting
> > looked like they were caused by VRPN version mismatches, so this may not
> > even be the current source of your problems.  My guess, from all of those
> > connection failure messages is that your firewall was, or is, interfering
> > with the socket connections between VMD and the VRPN server.  To check
> > whether or not the VRPN server is configured correctly, I would recommend
> > that you try and run the VRPN "printvals" program and verify that you're
> > able to get it to connect to the VRPN server, independent of VMD.
> > If the printvals program doesn't work, then you must still have some
> > configuration problem or some kind of network problem or firewall issue.
> > If on the other hand, the printvals program works fine, then probably you
> > are having problems due to a .vmdsensors config issue, or the VRPN
> > versions being mismatched, etc.  Try the VRPN printvals tests and let
> > me know what happens.
> >
> > Cheers,
> >  John Stone
> >  vmd_at_ks.uiuc.edu
> >
>
>
> --
> Dr. Axel Kohlmeyer    akohlmey_at_gmail.com
> http://sites.google.com/site/akohlmey/
>
> Institute for Computational Molecular Science
> Temple University, Philadelphia PA, USA. 

-- 
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349
  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078