VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Tue Jan 27 2004 - 16:31:56 CST

Steve,
  Aha, I'm glad we straightened this out. I've had a couple other people
get confused about this. Forwarding your X display connection to
a remote machine causes _all_ of the rendering to be done on the remote
machine rather than on the machine your're running VMD on.

So, if you are running VMD on your SGI but displaying on Linux, it is the
Linux machine's OpenGL renderer that is being used, not the SGI. Thus, if
you remote display to a Linux machine that has no hardware acceleration,
not only will it not run faster, but your performance will actually
be worse than if you just ran VMD locally on the Linux machine.

I normally recommend that people NOT run VMD remotely unless they
know what they are doing and have a specific need to do so, as this
is a common point of confusion.

Just for future reference:
  Anyone that needs help remote display VMD probably shouldn't
be doing it in the first place. :-)
There are lots of reasons not to do it, only one or two reasons to do it.

Thanks,
  John Stone
  vmd_at_ks.uiuc.edu

On Tue, Jan 27, 2004 at 05:23:35PM -0500, Steven F. Killen wrote:
> Yes, that's exactly it; the Mesa libs are correctly loading on the Linux
> box. I was under the impression that it would run on the IRIX box
> and just forward the graphics--shows how little I know about X forwarding
> over SSH. :)
>
> Thanks for bearing with me.
>
> On Tue, 27 Jan 2004, John Stone wrote:
>
> >
> > Steve,
> > Send me that actual startup message you're getting that says Mesa,
> > I want to see the actual text VMD outputs. From what you say, it almost
> > sounds to me like you're trying to remote display from an SGI to a Linux
> > box or something like that. If that's in fact what you're doing, then
> > the problem isn't with the SGI at all, its your linux box that's the problem.
> > The only place where VMD would print something about Mesa would be from
> > the OpenGL renderer string, and that comes from the machine console that
> > you're actually displaying to, not the machine you're running from, so
> > if you're remote displaying VMD to a different machine, that is the source
> > of your problems and you'll need to check the machine that you're actually
> > displaying on.
> >
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> > On Tue, Jan 27, 2004 at 04:02:09PM -0500, Steven F. Killen wrote:
> > > On Tue, 27 Jan 2004, John Stone wrote:
> > >
> > > > Steve,
> > > > Send me the output of this command from your SGI, so I can see what your
> > > > VMD is being dynamically linked against:
> > >
> > > irix2 /usr/tmp/build/vmd/IRIX6_64/lib/vmd % /usr/bin/elfdump -Dl
> > > vmd_IRIX6_64
> > >
> > > vmd_IRIX6_64:
> > >
> > > **** MIPS LIBLIST INFORMATION ****
> > > .liblist :
> > > [INDEX] Timestamp Checksum Flags Name
> > > Version
> > > [1] Apr 14 13:39:23 2002 0x5cf54b13 ----- libGL.so sgi1.0
> > > [2] Apr 14 13:39:34 2002 0xb1ced9e1 ----- libGLU.so sgi1.0
> > > [3] Apr 14 13:38:10 2002 0xba1b8c94 ----- libXi.so sgi1.0
> > > [4] Apr 14 13:37:17 2002 0x55e75e88 ----- libpthread.so sgi1.0
> > > [5] Apr 14 13:37:15 2002 0xb08e1179 ----- libX11.so.1 sgi1.0
> > > [6] Apr 14 13:37:22 2002 0x553d297d ----- libm.so sgi1.0
> > > [7] Apr 14 13:36:47 2002 0x6b322131 ----- libCsup.so sgi1.0
> > > [8] Apr 14 13:37:46 2002 0x2d614afa ----- libC.so.2 sgi2.0
> > > [9] Apr 14 13:36:48 2002 0xe58c9b39 ----- libc.so.1 sgi1.0
> > >
> > >
> > > It looks exactly like the output you listed. However, when I start up
> > > vmd, the little console window reports that it's using Mesa still.
> > >
> > > --
> > > Steve Killen <skille1_at_umbc.edu>
> >
> > --
> > 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
> >
>
> --
> Steve Killen <skille1_at_umbc.edu> 

-- 
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