[.net] Bones and destructable geometry C#/DX9

Started by
2 comments, last by EricHype1 18 years, 2 months ago
Hi guys, I've been mulling over this problem in my head for a couple days, and I can't come up with a good solution of how to do it, so I'm looking for some insight. I'd like to have movable characters that have the ability to have parts of them foricbly removed in the game I'm making (involves swordfighting). What I can't figure out is if I should: A) create each piece of the person as it's own seperate Mesh, and somehow map bones on to this (I'd have no idea how to map animation sets on to something like this) B) have my mesh be normal, but figure out how to hide parts of the mesh that occur after/below the "cut point" and generate a new mesh seperate from the player of his arm or whatever that got cut off and give that new dismembered arm it's own physics. (Not sure how I could dynamically load a part of a mesh. Any takers? ~Eric
Advertisement
You could look into Constructive Solid Geometry (CSG) and use that approach to take the intersection point of a sword and chop two objects from one. Then apply physics to the two meshes left over afterwards. Obviously throw some particle effects in there for blood as an added effect, though for smoother pre-scripted animation you can just go with multiple subsets in a mesh and have specific "splitting" points.

~Graham
After looking into the CSG stuff, that's not exactly what I'd want to do, thanks for the point in that direction though. It's definately some interesting stuff.

If I have a pre-defined number of cut points, say:
wrist, elbow, shoulder for 2 arms
ankle, knee, hip for 2 legs
head, head (vertical)
torso (horizontal), torso (Vertical)

thats a lot of meshes to carry around in memory for each character, and I don't want to be loading them during combat. What do you think of the idea of having the normal mesh with all the animations and soforth with "puppet" copy pieces overlayed over the normal mesh? That way I'd have for example a forearm attached to the whole person mesh, and the forearm "piece" mesh as a child of it. Then when an elbow cut happens, I'd have to hide the parent forearm and seperate the child "piece mesh" to now be it's own object.

Do you think that might be more efficient?
*Accidental double post that's me above*

This topic is closed to new replies.

Advertisement