Full Version: IK leg setup for spider ... need help and suggestions
Hey guys, I got a rigging question. I'm trying to rig a spider character, and I'm having a problem with an IK solver for the leg.is there a way that some bones down the chain would be made to be more rigid than others, like if there were 5 bones and they all give together 100% contribution towards the tip matching the end effector location, could it be made that based on an attribute, one joint could give 5%, and the rest of it's rotational duty be transferred to the other, or better yet specific other joints? On top of that, if anyone can give me a tip about rigging a leg that does not sit on one plane, like ideally for rigging you would want all the joints in a leg to be aligned on an arbitrary plane, but in this case the anatomy is such that the leg segments don't quite align even in a default/ neutral state. Any info or feedback would be super useful, thanks!
Hi, … er … Goran?
If you want to “weight” the bones in an IK setup in Houdini you most probably want to use an IK Setup with Constraints. That will, out of the box, allow you to use the “dampening” values and limits on angles.
Unfortunately, the name-value-logic in Houdini is counter-intuitive. A “dampening” factor of 0 in a bone (inside an IK chain) actually is “full dampening” (it does not contribute to the solution, so it should be named “contribution”, not “dampening”). A dampening value of 1 means “this bone can be fully rotated to solve”.
The angle limitations are to be understood as limits in “bone-world”. Again counter-intuitively the “dampening angle” only has a single value that you would THINK applies to the “core angle within the limits”, but it does not seem to behave like that. I tend to not use the dampening angle and it accompanying rollof, since both just do not give me anything near to what I expect them to give.
As for the angles of the bones in a leg - I would think that outside very old fashioned simplistic games no “realistic” rig would really have all bones in an extremity sitting on the same virtual plane. You will want to take care of “rest angles” so that the bones “know” where they belong when the IK solver has nothing to do - Houdini's documentation in that area is misty-foggy-vague at best, the best you can do is really to set up a 4-5 bones chain and experiment with what influences the chain's behaviour.
Sorry for being of little help only - I am currently struggling with Houdini's IK logic myself
Marc
If you want to “weight” the bones in an IK setup in Houdini you most probably want to use an IK Setup with Constraints. That will, out of the box, allow you to use the “dampening” values and limits on angles.
Unfortunately, the name-value-logic in Houdini is counter-intuitive. A “dampening” factor of 0 in a bone (inside an IK chain) actually is “full dampening” (it does not contribute to the solution, so it should be named “contribution”, not “dampening”). A dampening value of 1 means “this bone can be fully rotated to solve”.
The angle limitations are to be understood as limits in “bone-world”. Again counter-intuitively the “dampening angle” only has a single value that you would THINK applies to the “core angle within the limits”, but it does not seem to behave like that. I tend to not use the dampening angle and it accompanying rollof, since both just do not give me anything near to what I expect them to give.
As for the angles of the bones in a leg - I would think that outside very old fashioned simplistic games no “realistic” rig would really have all bones in an extremity sitting on the same virtual plane. You will want to take care of “rest angles” so that the bones “know” where they belong when the IK solver has nothing to do - Houdini's documentation in that area is misty-foggy-vague at best, the best you can do is really to set up a 4-5 bones chain and experiment with what influences the chain's behaviour.
Sorry for being of little help only - I am currently struggling with Houdini's IK logic myself
Marc