Talk:Ifd:Second LexiCon-BARBi Unification workshop

From dev.ifd-library.org

Jump to: navigation, search

On the topic of (parametric) geometric model inclusions

In this section OpenGL is being proposed as a possible means to store parametric 3D geometries in IFD. U3D is being mentioned as another candidate means. My two cents:

  1. Capturing truly parametric geometries is a highly non-trivial field
  2. many approaches to do this have been worked upon by hundreds of engineers since the early days of CAD. Many promising ones have died a painful death along the way.
  3. We do not have the (human and knowledge) resources to come up with s.th. 'from scratch on our own within a year
  4. bad news:
    1. 'pure and naked' OpenGL is definetly not an option. It's not even somewhere near a file format or storage, it's simply "a software interface to graphics hardware" ( from the OpenGL spec ).
    2. allthough there are some graphic primitives defined in the OpenGL Utility Toolkit (GLUT) like box, cone etc., it's still nowhere near what is needed for parametric objects.
  5. good news: there are many approaches, existing initiatives and people working on this:
  • There are vendor specific implementations of parametric objects, like [Graphisoft's GDL] and others
    • Plus:
      • it works
    • Minus:
      • vendor-specific
      • outdated
      • pretty horrible to code if you ask me
  • There is X3D and VRML97 who have a very powerful PROTOTYPEing mechanism (that can be connected to ECMAsrcripting, hence making things like DoorWidth=DoorHeight/2 and more complex procedural things possible)
    • Plus:
      • it's XML with an optional binary compressed version.
      • it's open.
      • many existing tools, libraries, viewers.
    • Minus:
      • not much support for standard CAD vendors
  • Somthing is happening within STEP / ISO 10303 : part 108: "Parameterization and constraints for explicit geometric product models". I have no idea, where they currently are, but maybe it's worth having a closer look at this and maybe try to get in touch with the guys (Michael J. Pratt seems to be the driving force here, I will browse through some of his publications. A free available paper (thoug somewhat old) covering the basics is here. If you have access to Elsevier, there is a recent one here).

--Jakob.Beetz 17:24, 4 May 2006 (CEST)

Retrieved from "http://dev.ifd-library.org/index.php/Talk:Ifd:Second_LexiCon-BARBi_Unification_workshop"