Jump to content
  • Advertisement
Sign in to follow this  

Optimizing Isometric Mesh of Parametric Surface

This topic is 3096 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Until now I always sampled my parametric surfaces on a grid. However, while the control points for parametric surfaces (i.e. NURBS) usually lie on a grid, the surface UVs of the sampled points don't have to. The problem with grids is in thin areas or pole areas where the polygons converge and many are wasted: It is preferable to use "isometric" meshing - where the triangles attempt to be the same size. A simple, fast to execute and "fast" to implement (40 hours) approach (when you have parametric surfaces) is to simply prorate each slice by the radius and then perform N to M stitching on each slice to the neighbor: This approach uses between 25% and 75% of the polygons of the grid approach. However the problem is that all of the slices start at 12:00 and in extreme cases it creates this effect: I'm wondering how to balance it? Some ideas: 1) Grid based - the surface UV's should look like graph paper that was cut with scissors, where each vertex snaps to the grid, and portions are on the edges. 2) Padding approach - pad the edges to balance things, introducing smaller triangles on the edges of open areas. On closed areas, this gave the polar regions a spiraled look that was an improvement. I presume that ideally I would postfilter the mesh and allow points to to lie off of the slice boundaries themselves?

Share this post

Link to post
Share on other sites
Sign in to follow this  

  • Advertisement

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

Participate in the game development conversation and more when you create an account on GameDev.net!

Sign me up!