VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Thu Jul 17 2003 - 11:21:22 CDT

Hi,
  The STL format used by 3-D printers has a number of very strict
requirements which VMD cannot promise to meet any but the MSMS representation. The STL spec requires that surfaces not have self-intersections, that
neighboring triangles share vertices, that they don't have T-style
triangle joints, etc. The only representation that really follows
most/all of the rules is the MSMS representation. If you want to
export other VMD representations as STL, I'll wish you "good luck" but
I make no promises that your 3-D printer will accept them. The latest
round of 3-D printers have much more advanced software which is able
to cope with some of the issues I listed above, but even these may only
do a so-so job of creating a solid model from the thin geometry one gets
using any of the reps other than MSMS.

While it is true that the newer versions of VMD have made significant
improvements to the range of geometry that the STL export will output,
because the actual representations themselves do no gaurantee
non-self-intersection etc, there's no way for the STL export to
generate a STL file that's 100% compliant. The STL format is
extremely limited in what it allows, so its not surprising to me
that it gets upset when you export anything other than an MSMS
representation. We added code to allow users to export more geometry
than in past versions at the request of the VMD user community, but
as I said above, due to the limitations of STL, I can't make any
promises that any molecular structure representation other than
an MSMS surface will result in an STL file that is actually printable
by a specific printer.

If you want predicatable results, stick to printing MSMS surfaces and you
should be fine.

Thanks,
  John Stone
  vmd_at_ks.uiuc.edu

On Thu, Jul 17, 2003 at 01:07:02PM -0300, alarcon_at_lncc.br wrote:
>
> Dear all
>
> When exporting a pdb file (structure) in a STL file format, only the CA trace,
> surface and ball-stick representation works fine. Unfortunately, my demand is
> for a cartoon representation (helix as cylinders, strands as arrows, coils as
> coils).
> This representation always have several defects, mainly the not construction of
> the cylinders and coils themselves (arrows have a lot of triangles with vertices
> not common to neighbour triangles and other misplacements).
> Is there some manner to fix that or other software to acomplish this ?
> I was told that the newer version of VMD could fix that.
>
> Thanks in advance
>
> F. Alarcon
> www.brgene.lncc.br
>
>
> -------------------------------------------------
> This mail sent through IMP: http://horde.org/imp/ 

-- 
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