On a fun note for me, the arm grabbing motion shown at the end makes me unreasonably happy, wasn't expecting it, keeping it in now I've seen it.

I just had to step away from the project BEFORE I coded the puppets twisting priority and applying system and not AFTER didn't I.

Long and short of it, it's nice that most of the errors I'm hitting so far, are things I predicted might happen, and already had plans for.

Makes me almost feel like I know what I'm doing here

Yep... turned out my 'Apply Child Movement' function was the cause of the extra torso wobble.

I'll disable it for now, but fingers crossed the 'Axis Drag' function I've had planned anyway, should fix that when I code it in later.

Anyway, need to debug if the extra motion is an issue in the twisting system, or some kind of compounding loop from the parts being dragged along, feeding extra force to the hips.

If it's the latter, I might be able to leave it for now, since better child motion was on my fix list anyway.

Best advice I can give, is just go for it, you'd be surprised how much you can get, with even just two spheres for the torso, and some basic IK code for the arms and legs.

So long as you don't mind the motion being a bit comedic as you start out.
www.youtube.com/watch?v=hQub...

Can't point to any I'm afraid. When I started researching this there wasn't much out there so I've just been working it out as I go

However, I know there are more and more videos on things like active ragdolls, and procedural systems that could help for those wanting to start doing things like this

A public note to self.

Passive forces, aka gravity, momentum, should have less effect on the posing of the torso when it's connected limbs are anchored and more when they are not.

Legs for the hips, arms for the chest. Would make climbing work even better, or just wall leaning I guess.

More testing and just by enabling the chest point statically between the two hands, I suddenly get so much swinging in the motion!!!

And again, barely any work for a system to implement, this is shaping out to be pretty nice
#proceduralAnimation #gamedev #videogames

Quite surprised by how well the legs hold out during this motion.

And a extra plus for me, the tension they're missing on the forward swing, is something I already predicted a while back, and already have plans in the work for how to solve... so this was validating.

And for a non technical note on this video.

I like how they bounce, it makes me laugh. Here's hoping I can get that kind of motion to be intuitive.

This is basically were I go from working on a limp actor, to a now a semi-smart directed one.

The hope is by the time I'm done, I can direct the pose by adjusting all the inputs, or only one, and still have the rest line up properly.

Then while I'm writing future dreams, if I can get a sword system that can bounce off, get stuck in, or slide off things using this motion system...

...Then I can imagine doing some fun stuff with this if I take some of the logic from past combat systems I've done.

For insight to longer term plans, it's then sword to scene collision, playing with losing control and momentum on bigger body movements.

Once I have those in, it'll then probably be time to start adding proper state managers to handle intelligent actor to scene interactions and player input.

My aim for the next step and those after, is to develop the motion system by making playable examples, now the foundation is sturdy enough.

So this sprint target
- Moving and rotatable floating body that effect pose.
- Dynamic sword swinging motion.
- Basic wall collision.