Re: [opendx-users] elliptical tubes

[Chris Pelkie]

Actually, Richard Gillilan implemented this in his module FancyRibbon which is part of the Cornell Chemistry package. I've lost track of how that is distributed. Is it linked from opendx.org? The only trick of course is that you must be able to do a rather long compile/install on your local host and I'm not sure what hosts it's compatible with other than SGI and Mac OS X. The appropriate config tools are part of the package, I think, but I'm NOT the one to ask for installation instructions.

FancyRibbon splines the 'backbone' positions and permits elliptical profiles, parameterized by the normals and binormals at each position. He wrote it for doing proteins, but I've used it for reconstructed plant venous systems (from slices), vortex shedding vizzes, and other stuff. Much nicer than Tube if you can get it installed on your platform.


On Friday, May 28, 2004, at 03:00 America/New_York, Allen H. Nugent wrote:

Marc,

- third and last, could I have a tube with an elliptical crossection, where
the parameters for the ellipse come from some of the data columns of my
input data file ?

What you're talking about is called a "hyperstreamline" (see, for example: Delmarcelle T, Hesselink L: Visualizing second-order tensor fields with hyperstreamlines. IEEE Computer Graphics & Applications 13(4):25-33 (1993).

We'll have to see about getting this feature into openDX, someday!

Regards,


Allen H. Nugent
Graduate School of Biomedical Engineering
University of New South Wales
Sydney NSW 2052 Australia
Tel: +61 2 9385 3916 Fax: +61 2 9663 2108

_______________________________
Chris Pelkie
Scientific Visualization Producer
622 Rhodes Hall, Cornell Theory Center
Ithaca, NY 14853