View Full Version : use of bones and skin to avoid vert collisions

03 March 2006, 06:53 PM
I've been watching Paul's third DVD on rigging and in it, he uses stretchy bones to push verts out when two bones bend towards each other. Very interesting as I was always referring to setting up Morph Angle Deformer gizmos within 3dsmax's skin modifier.

In cases where the collision of vertices occur as bones bend towards eachother (such areas are the elbow and knee joints)....would it be smart (verus using Morph Angle Deformer Gizmos) to incorporate the use of small stretchy bones that rotated slightly in the opposite direction to their paranted bones and skin the colliding verts to these smaller bones. These child bones would be pivoted at the same location as their parent bone's pivot, just rotating slightly in the opposite direction as the parent bone rotates perhaps using Orientation Contraints between say the femur and tibia bone (knee example). Then in these tight areas (such as the elbow and knee), the verts would be following the child bone direction and thus be moving away from each other rather than towards and into each other?

Sound strange? Any thoughts or suggestions on how I can improve my thought within 3dsmax?

I may not be explaining it as I'm envisioning it.

03 March 2006, 03:27 PM
Not sure that this would work but them I'm not completely sure what it is you explained. Only way to find out is to try it and see what happens. That is how I have learned everything in rigging I think.

I also wouldn't use the gizmos in Skin but instead use Skin Morph as this is the best tool in the world for correcting joints, I'm using it right now actualy.

03 March 2006, 02:49 PM
the end user is on 3dsmax v6 so I'm looking for a solution using skin weights and bones while staying away from Skin Morph Angle Gizmos (I wanted to see if it was achievable without the use of the Skin's Angle Gizmos).

I created a small bone that was linked to the tibia (shin) bone so that it's pivot ended up at the back of the knee joint. the small bone was 'pose-angled' near to a 45 degree angle from it's parent bone's x axis (rear of the knee area). This is the default pose when the character is standing up.

I then added a reaction controller to the small bone's x axis so that when the tibia bone's x angle rotated from the posed values through to where the tibia is fully bent (visually around 30 degrees at the shin bone's pivot), reaction controller would then rotate the small bone's x value so that the small bone ends up perpendicular to the femur bone.

Blended the verts' weight values behind the knee to this small bone along with the tibia and femur.

Question: are reaction controllers within 3dsmax slower than say an alternative like a script controller ? I thought of orientation contraints but that would rotate on all axis and this 'hypothesis' was only looking to rotate on a single axis.

03 March 2006, 05:36 PM
Reaction controllers I think are slower then others but it all depends on how you are using them. I only use them if I need a function curve to control the value. Use other methods if you can.

CGTalk Moderation
03 March 2006, 05:36 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.