• Advertisement
Sign in to follow this  

Hierarchical scene - node orientation

This topic is 4169 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

Hi, I've just implemented a math library for handling objects position&orientation in 3d space (matrices + vectors + quaternions = that's all now). Now, I want to represent a hierarchical scene. Basic idea of nodes (father/son & siblings) sounds good to me, but what if it comes to handling input? A mouse picking, based on ray-triangle test has to be implemented. So the basic idea was to "just store the inversed model-view matrix" for every model. And here the doubts comes: Should I handle a global "absoulte" coordinate system, and store (for every node) a pair matrix & inverse to allow direct absoulte<->local transformations? Or maybe, I should just to store father<->son matrices? Which one is better for space-partitioning algorithms, like octrees? Should these matrices be precalculated, and then transformed every frame (when animation occurs) or better should they be built every frame for a pair <vector, quaternion> ? Yeah, when it comes to designing, it's all about experience, isn't it?

Share this post


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

  • Advertisement