Youre talking about two different things; IK and Match Goal Orientation. The IK should bake fine using motion baker or Dstorms IK Baker plugin. I would use IKBaker because it just alot easier. Match Goal wont bake which requires that 3rd party match goal plugin.
Heres something else you can try:
Instead of using match goal orientation use Orient Contraints. That will give you the same result as match goal except that it will bake by using motion baker.
Might give that a go then. Never really used orient constraints as I am not a great animator myself and have just dod bits here and there. Am looking to get into properly so thanks for info. Will have a look and see if I can find it
The best way is using the BakeIKtoKey plugin. You do have to remember it only affects items which have IK set as the controller for some of their channels, so if you're using Simple Orient Constraint on the foot, (to get match goal orientation functionality and still be able to bake it) you have to set the foot's channel controllers to IK. You also have to bear in mind it'll change those channels back to FK when it's done, so backup your carefully set up scene first
You can use Motion baker instead, applying it to each bone and setting it to use original channels and only set it to bake the channels which use IK. This is nice because it updates key frames as you scroll through them if you make changes to the motion later on, but is much more manual to set up and turn on and off and key reduce. So BakeIktoKeys is better if you remember to keep a scene backup.
Here is the rig as promised. The reason for having the bones set up this way, is that when a model is parented to the bones' position and rotation, you will have more control over what is going on. I am also planning a ragdoll system for a rig like this, and then I definately need a model with "predictable" bones tranformations.
Wow, this is very interesting. I thought the bones had to be connected, as in a conventional bones skeletion, for it to work in W3D format. Why did you choose this way to place the bones? Is it so you can easily use the rig with multiple models? I see that you just have to have the bones positioned on the joints, which might make it easier. Or is it simply that you like to see the bones more clearly? Thanks for sharing - I've found your posts extremely helpful with my learning curve on W3D.
Glad to be of help, I dont gain anything by keeping it to myself "sharing knowledge gooood"
Yes, that is the essence of why I chose to do it this way. The only thing I have to do now, after creating dozens of animations and exporting them individually as animations, is to export the models at the neutral pose and apply the animation from previous W3D exports through code. Since the animations are all based on rotation only, it does not matter if the animated character is 1/10th of the size (or different proportions) than the original. But if you are making a troll, i guess would might want to custom make some animations for it, as it is walking quite differently than the average human. Before this will work 100% I have to take this rig one step further, and add a "toes" bone on each side, and these bones will be the ones deciding the up/down movement of the character. That means I will remove the up down movement from the root bone in the animation, and steer the up down movement based on the position the toes bone have to the ground.
Check out the "toon" line around the character, its basically a smooth scaled version of the same model with its polygons inverted and black surface
Oh and on the issue of connected bones, they are really just bones pointing at each other with the rest length long enough so they look connected. These bones have the exact same parenting, but no "pointing" at eachother. But I must admit, it looks nicer when they are connected
Ha ha! Awesome.
I couldn't figure out how you got the toon line. Very very nice. Isn't there a toon shader in Shockwave? I haven't tried it.
In case you hadn't guessed, I'm working on a figure for shockwave myself. It's quite experimental at this point (and I'm too embarrassed to post it) and I'm only just learning lightwave and shockwave for the first time, but once it works a little better I'll share. I'm aiming for something a little more detailed, with fingers and facial expressions. 100% weight maps aren't really ideal but I seem to be able to get some pretty good results.