Animation - Keyframe vs Skin-and-bone
The 'keyframe' being referred to in the post above is 'vertex key framing' as each vertex position is stored at each key frame.
Both skeletal and vertex-blend animations are 'key frame' animations, it's just how the data is stored and the resulting mesh is constructed which changes between them.
Both skeletal and vertex-blend animations are 'key frame' animations, it's just how the data is stored and the resulting mesh is constructed which changes between them.
Yeah, there's a lot of incorrect terminology being used in this thread.
What's basically being discussed is morph targets vs skeletal animation.
Key-framing is a completely different concept, which can be used with almost every form of animation, including both of the above (both of the above can also be used without key-framing)...
In a general sense, morph-targets are require more memory, whereas skeletal requires more computation power.
Skeletal animation can be extended to use the "Skinning" technique -- which is where each vertex can be attached to more than one bone to get more natural movements around joint areas.
Skinning only became really popular once vertex shaders came on the scene and gave us that level of computation power.
Most things can be animated via either, but fairly rigid objects (like skeletons) work well with skinning, while fairly soft objects (like faces) work well with morph-targets.
Modern games often use both techniques at the same time (e.g. morphing a character's face, while also using a skeleton for their body).
What's basically being discussed is morph targets vs skeletal animation.
Key-framing is a completely different concept, which can be used with almost every form of animation, including both of the above (both of the above can also be used without key-framing)...
In a general sense, morph-targets are require more memory, whereas skeletal requires more computation power.
Skeletal animation can be extended to use the "Skinning" technique -- which is where each vertex can be attached to more than one bone to get more natural movements around joint areas.
Skinning only became really popular once vertex shaders came on the scene and gave us that level of computation power.
Most things can be animated via either, but fairly rigid objects (like skeletons) work well with skinning, while fairly soft objects (like faces) work well with morph-targets.
Modern games often use both techniques at the same time (e.g. morphing a character's face, while also using a skeleton for their body).
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
