Re: [brlcad-devel] Questions about metaball

[p. <284...@qq...>]

Hi, 


> I think this is an accurate summary.  The only papers I can find that
might pertain to this topic offhand are:

> http://nis-lab.is.s.u-tokyo.ac.jp/nis/cdrom/eg/Eurographics94beziercolor.pdf

> http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.14.2762

 I have read these two papers. But I could not find the information to answer my question. The first one (A method for displaying metaballs by using Bezier Clipping) seems to be related to intersecting a ray, using Bezier Clipping to solve the root of the equation to find the intersection, while the second one (NURBS based B-rep Models for Macromolecules
and their Properties) studies with macromolecules, which are quite different from the metaball models. For example, if we put two balls in the same position with the same radius in the metaball model, the combination should be a ball bigger than a single one. But the macro-molecule model cannot do this. In the macro-molecule model, we draw all single spheres (atoms) first, and then the contact surfaces. So I wonder whether they are suitable for this project.


 I have sent the list of primitives with whether their b-rep conversions have been implemented. Only a few of them are missing, and metaball is one among them. But I found it hard to get started. I really want some advises and suggestions.


Chees!
Wu