From: John Stone (
Date: Fri Nov 20 2020 - 14:19:24 CST


On Fri, Nov 20, 2020 at 02:52:23PM -0500, Lothar Esser wrote:
> Hi, John,
> thanks for your super fast reply. A few days ago, I had a sent a very
> detailed account of the problem to [1] but never heard a
> thing.

I've been totally snowed under by almost a thousand emails in the last
four days or so. I'll find it eventually...

> In the meantime I did make some progress. It seems you are saying
> tachyon should render colored surfaces and only in my unusual case(s) it
> seems to fail. The example that I provided back then is simply crambin (I
> have a much more complicated molecule but here crambin 1crn.pdb failed
> already).

It shouldn't even matter what particular molecule is being used.
I'll just have to look carefully at your saved state and see what
might be happening.

> BTW I think you are probably right: The new APBS required a version of gcc
> that I did not have and had to compile. One library ended up in

That is the most likely cause of the behavior you describe that I could
come up with. I'm not aware of any other reason why it wouldn't work.

> Okay, I am attaching one example. The my_vmdscene.tga is what Tachyon
> produces for me.

Thanks I'll get back to you after I have a look.


> Thanks,
> Lothar
> On Fri, Nov 20, 2020 at 1:58 PM John Stone <[2]> wrote:
> Hi Lothar,
> APBSRun tries to locate APBS using your default command
> search path, as it exists when VMD is launched. Since APBS
> may depend on shared libraries set in LD_LIBRARY_PATH, there's
> an extra potential complexity, because recent Linux distros
> have decided to clear out LD_LIBRARY_PATH for shell child
> processes as a (in my view bizarre) form of extra security.
> The upshot of this is that if you require a special setting of
> LD_LIBRARY_PATH for ABPS, that may be the source of the problem
> and not its inclusion in the default command search path.
> Regarding your difficulty with Tachyon, if you can provide
> me with a test case that reproduces the behavior your describe,
> that would be the best way forward. Although it "works for me",
> you may have found a strange case where my code is misbehaving.
> I would want a tar.gz archive containing the electrostatic map,
> your VMD saved state file, and the exact "render" command you
> ran to generate the image(s) that are problematic. If you're
> using the GUI to launch Tachyon, you can use the "logfile console"
> command to get VMD to tell you what render command it is running
> in response to your mouse actions.
> Best regards,
> John Stone
> [3]
> On Fri, Nov 20, 2020 at 01:30:57PM -0500, Lothar Esser wrote:
> > Hi,
> > because of the excellent visual representation of surfaces when
> ambient
> > occlusion is used, I am planning to render a surface with
> electrostatic
> > potential in Tachyon (in-memory). However, tachyon only produces black
> and
> > white surfaces. That is in contrast to colored surfaces say colored by
> atom
> > types. So this is surprising.
> > I am using 1.9.4a48 on CentOS7.
> >
> > Also, I noticed that whenever I use ambient occlusion, regardless of
> > whether I turn on or off Shadows in Graphic options, it renders the
> scene
> > *with* shadows. Previous versions of vmd did not do that as far as I
> > remember.
> >
> > A smaller side issue is that vmd does not execute APBS even though it
> is in
> > the path. After running APBS (local) it displays immediately: APBSRun:
> > output files missing or unreadable.
> > Obviously it does not even try to execute it. However the [4]
> file can
> > be run on the commandline with apbs [5] and the pot.dx file
> copied to
> > the working directory.
> > Not that important but, if someone knows whether this is a permission
> issue
> > or something else, I would like to know how to fix it.
> >
> > The major issue for me is that Tachyon does not render surfaces
> colored by
> > (APBS) electrostatic potential and to a much lesser degree the problem
> that
> > shadows are always on even when they are turned off.
> >
> > Any advice is appreciated.
> >
> > Lothar
> --
> NIH Center for Macromolecular Modeling and Bioinformatics
> Beckman Institute for Advanced Science and Technology
> University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> [6] Phone: 217-244-3349
> [7]
> References
> Visible links
> 1.
> 2.
> 3.
> 4.
> 5.
> 6.
> 7.

NIH Center for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801           Phone: 217-244-3349