From: John Stone (
Date: Thu Aug 02 2018 - 11:58:08 CDT

  Sorry for the slow response, I'm still digging out of a pile of emails
that accumulated while I was away from the office for a few weeks.
Did you try using the built-in version of Tachyon I ship with VMD itself?

Tachyon always had DoF, but the parameterization has changed somewhat from
the old versions to now, and the way VMD talks to Tachyon has changed also.

I haven't made the bleeding edge version of the Tachyon code available yet
because I've been revising its API to allow it to support interactive ray
tracing for machines that lack GPUs, and that aren't supported by OSPRay.

If you have a problem that crops up with the test VMD build (and its built-in
version of Tachyon) on the web site, then let me know and I will address it.
If TachyonInternal works but the external does not, that may be a scene
file generation issue (two different code paths). In any case, it should
be easy to resolve any loose ends that may arise here.


On Wed, Aug 01, 2018 at 10:48:45AM +0200, Peter Kroon wrote:
> Hi all!
> I sent this mail earlier, but I think it got buried in everyone's
> holidays. Does anyone who's just come back from their well-earned
> holidays and conferences have any insights?
> Peter
> On 09-07-18 16:36, Peter Kroon wrote:
> > Hello world!
> >
> >
> > I would like to turn a VMD scene into a pretty picture. I want this
> > picture to be high resolution (~3315x4000px), and using AO and Depth of
> > Field. I have tried the following, continually running into issues:
> >
> > 1) Use Tachyon-OptiX: I can't put the desired resolution. If I do, it
> > fails immediately with a memory error. See also
> >
> >
> > 2) Use Tachyon internal: Creates a super fuzzy picture, probably because
> > the meaning of 'Aperture' changed from 'x' to '1/x'.
> >
> > 3) Use external Tachyon renderer: If I fix the 'Aperture'; and changing
> > FocalDist to FocalLength I *almost* get the picture I want, but it seems
> > the FocalLength option is simply ignored (see also
> > This
> > happens with both 0.98.9 and 0.99b6.
> >
> > Does anyone have anything else I can try? I build all the software from
> > source, in case it matters.
> >
> >
> > Peter
> >

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