I’m particularly interested in how hair can dynamically respond to character movements such as walking, running, and jumping.
Has anyone here worked with Havok-based physics for this purpose? I would appreciate any insights, experiences, or examples you could share.
cc @Cedric
I would say it depends 
If you group hair and have a few lash, then Havok might be OK.
For 100s, I would simulate on the GPU.
How does it work generally in the industry? When designing a 3D model, do artists add a few bones to the hair, or is each part (let’s assume the hair is constructed from 3-4 parts) wrapped with a physics body? I’m looking for something very simple, like the character from Zelda, whose hair consists of 4-5 sections.
It also depends on how you update the mesh. If it’s skinned mesh, you have to update the skeleton bones and It can be a bit difficult with gltf.