good comments.
controlling real-time head/torso animations will probably require new engines, instead of engine mods.
And it won't be considered high priority to include such real-time functionality in the intial "engine-design" of a product, until everyone is already using it. sooo, this could be a long ways off.
just want to note the worst case scenario for a minute.
To get TrackIR built into Well-Known FPS games ASAP, we'll probably have to give up on the idea of constantly responsive head/torso animation.
I've read about new animation techniques (in the Unreal engine?) that blend between different predefined sequences on the fly, so that animators don't have to consider every situation. if this approach catches on, it seems like it might help us a lot.
http://www.ogre3d.org/index.php?option= ... &Itemid=62
http://www.unrealtechnology.com/flash/t ... /ue2.shtml (?)
http://www.epicgames.com/UnrealEngineNews.html
(cross fingers that this catches on quickly. hope hope)
(i do hope that by the end of the year we could at least have modders somehow remove the head from existing animations, and replace it with a head that moves in small amounts around a central neck point, based on TrackIR input. this small positioning modifier info could be passed over networks easily. it could scale from as little as "4 snap-to positions" to as much as "realtime insanely precise constantly" is that the technical term? heh. Predicting it wouldn't be necessary. Hickups wouldn't matter much because the rest of the body would be solid/predicted. I'm picturing it really just as an eye candy thing. a middle step)
i expect we will push for FPS devs to implement us in a subtle way to start. ie, you'll enjoy a little bit more immersion because the view moves a small amount away from your mouse in any direction/rotation. but it won't move enough to break the models and/or bounding boxes.
(ie you won't be able to look down enough to see the ends of your arm models and no body.
You won't be able to lean forward enough to poke through a wall).
For now, i don't think we want to risk scaring developers off by having TrackIR give it's users any big time advantage. I don't like the idea of castrating the device, now that it's so awesome, but asking the devs to totally rework their tools for animation, networking, and hit detection is A LOT TO ASK. (weeping).
we'll see though. anything can happen.
We're now beta testing some fps implementation mods, so i'd love to hear more ideas/thoughts about how to use TrackIR with Vector Expansion in a small way, that doesn't require crazy animations that are visible to other players.
(or talk about your dream implementation if you like. heh.)
[ March 30, 2005, 09:50 PM: Message edited by: NaturalPoint-Warren ]