... and have felt as if I had a firm grasp of forward and inverse kinematics. The discussion where I saw your comment on the subject was in reply to my C9 pro wishlist. It appeared to me that you tweak out the IK chain of the figure itself, rather than using the helpers… is that right? If so… I may try that approach instead - using the helpers only where needed.
I’m confused about what you’re asking. Here’s a very basic explanation of IK, maybe it will help.
When you reach for a cup of coffee on your desk, your brain tells your hand to go to where the cup is located. Your hand starts moving in that direction. Your forearm bone, shoulder bone, collar bone, and spine bones all follow along automatically to help the hand get there and maintain balance in your body so you don’t fall over. It’s automatic. The hand goes, the rest of the body follows along. That’s what’s called “inverse kinematics”.
However, first your brain has to tell the hand where to go. That’s the IK goal. It’s the goal for the IK chain.
You MUST have an IK goal to tell the IK chain where to go. The IK system includes a behaviour that “pulls” all of those bones and muscles, starting with the hand, in that direction. But without that goal, it wouldn’t work.
Now, if instead your brain tells you to pick up the cup, and you first have to figure out how to rotate your shoulder in that direction, then how to rotate your forearm, your wrist, your spine, etc., that’s different. That’s forward kinematics, and it’s what you do in any 3D app to pose your figure when IK isn’t applied.
So I don’t know what you mean by “tweak out the IK chain rather than using helpers”. There is no “rather than”. You need a goal. You can use a cube, or a sphere, or a sword, or another V4, or a target helper null, but you’ve gotta use something. Usually people use null objects (target helpers) because they’re more convenient.
And no, collision detection doesn’t do that.