Bakerman on 17/9/2009 at 16:04
I got the idea for this thread while reading the FP/TP poll. I just figured I wanted to discuss things separate from the issue of first person/third person, hence a separate thread.
I'll give you my perspective on this: I'm currently developing an FPS game, and right from the off I'm discarding the conventional FPS viewpoint epitomised in Halo, Half-Life, and yes, Thief: the floating camera with an arm attached. All my coding is going twards making each character move and animate realistically. Then I simply attach the camera to the character's eye node, and there you have it.
Now, how does this apply to Thiaf?
I've heard complaints about T3's 'body awareness' and people who want to return to the old floaty-camera style of control.
Why? In the Builder's name, why?
How is Thief's first-person perspective more immersive than that of Dark Messiah, or even Deadly Shadows? (Of course, I appreciate that it had leaning poblems, and that annoying swivel when you move after looking around - but they're not issues with body awareness.) How is it any better to feel like you don't exist than to be able to see your own body and feet?
Ostriig on 17/9/2009 at 16:14
Every first-person game should have full body awareness. It's not just about furthering immersion through "realism", but also a massive improvement for gameplay purposes.
Ostriig on 17/9/2009 at 16:52
Quote Posted by Wormrat
1) Seeing your body and feet is not the same as body awareness. The proper term for a camera mounted inside the player model should be "true first person."
Seeing your body translates to being aware of it, at least to some degree. Your "true first person" could mean a whole lot of different shit, what you're suggesting is an anatomically correct camera placement. Then again, this is all a matter of semantics and terribly irrelevant. The point of calling it "body awareness" is that you're given a direct visual cue on where your body, and more importantly, its collision element, is.
Quote:
2) Animations are limited. No one's even come up with a good way to add an "open door" animation that looks appropriate for all the possible orientations the player can have to the door. Combine this with
Thief-style frobbing, and you can see that body awareness doesn't easily map onto the actions the player is supposed to be taking. For example, I'd rather see nothing when I pickpocket someone and just accept that I don't have all skills my avatar has.
And I'd rather see my feet when I'm on a ledge. Or when I was jumping over laserbeams in DX.
What you're talking about has to do with arm animations, and the difficulty in translating those animations to connect to a variable 3D point. That
does not serve the gameplay purpose of body awareness, and it's simply an aesthetic matter which could have benefits in furthering immersion.
Furthermore, animations are a lot farther along now. Believable animations for picking up objects or opening doors are doable, though it would take a lot of effort, granted. At this point, Fallout 3 already uses IK in third-person mode to more accurately place your feet (and, by extension, bend your legs) on the varying height levels of whatever's under them.
Quote:
3) Similarly, body awareness rarely does a good job of simulating anything other than how your body can obstruct your view. Is it immersive to shuffle your feet around to see the floor, instead of just moving your foot like you would in real life? Even if you
could add such specific controls, it would be a pain in the butt.
See above - a visual cue to your exact position can be very useful. On top of that, what you're arguing out is sloppy implementations, not a problem with the concept itself. Implementing a simple hierarchy of animation for different body parts when turning in a fixed point would solve most of that.
Quote:
[...]
Body awareness in video games is basically just a big misnomer.
No, you're just limiting yourself to considering outdated or straight-out lazy implementations of the concept.
New Horizon on 17/9/2009 at 16:55
I wouldn't call body awareness 'true first person'. True first person would remove awareness that you're attached to anything at all. The only thing body makes me aware of is that it's not my own body up there doing things on the screen. The floating arms are just an extension of my hands that are already doing things on the keyboard and mouse to translate commands into on screen movement. Strap some motion sensors to my feet and then I'll have some connection to the rest of that body onscreen, until then...I want control rather than eye candy...thank you very much.
ZylonBane on 17/9/2009 at 17:03
Quote Posted by Bakerman
Why? In the Builder's name,
why?
Because in real life, the overwhelming majority of the time you're not thinking about your own body. "Body awareness", on the other hand, forces you to deal with a simulation of a body other than your own. Even the rudimentary nod to realism of head-bob is too much for some people... body awareness takes the same problem and makes it a hundred times worse.
The second issue is one of control. The player's control of their in-game avatar is necessarily highly abstract-- you're mapping an entire human being with thousands of individual muscles to a pointing device and a handful of buttons. The upshot of this is that, for all but the most trivial motions like "walk forward", the game simply doesn't have enough information to animate a fully modeled human in any sophisticated way. Look what happened to Deadly Shadows-- we ended up with constant uncommanded sidesteps, and ladders that you were bolted to and wouldn't even let you look more than a few degrees to either side. They made player agency a slave to the animation system.
The proprioceptive transparency of classic FPS games is, therefore, more immersive and provides superior control. Body awareness sucks. Deal with it.
Ostriig on 17/9/2009 at 18:37
You just quoted two examples and beat me to the third, Mirror's Edge. We're at the point where games can implement more complex physical models, and being able to discern visibly where you are, where your collision box is, is preferable, in my opinion, to having to appropriate the relation between that floating camera and an invisible collision element through, likely, guesses and experience. Especially since even after hours of play you can still find yourself making a mistake, tripping over a beam or knocking a chair over.
I don't know, that strikes me as a pretty useful thing to have and I still don't see the disadvantages.
Quote Posted by ZylonBane
"Body awareness", on the other hand, forces you to deal with a simulation of a body other than your own. Even the rudimentary nod to realism of head-bob is too much for some people... body awareness takes the same problem and makes it a hundred times worse.
The second issue is one of control.
[...]
I don't understand, what additional function or control do you see being added to the game by introducing a visual representation of your body? As far as I can tell, the collision model stays the same, as does the way you're controlling your character. We're simply talking about putting in a mesh and an admittedly comprehensive set of animations and their automated triggers. Yes, we can't animate a fully functional human based off the keyboard input, but we don't have to, it goes the other way around. And providing fairly fluid and accurate animations for the drastically limited positions that we can and are required to put a character in with the current control schemes is doable.
ZylonBane on 17/9/2009 at 18:47
Quote Posted by Ostriig
I don't understand, what additional function or control do you see being added to the game by introducing a visual representation of your body?
Try actually reading the paragraph you almost quoted. It answers all your questions.
Now here's a question for you: In the good Thief games, when on a ladder you can move all around it and look in any direction. How would that be animated with a body awareness model?
Namdrol on 17/9/2009 at 19:02
For me the closest analogue to a truly immersive game is either reading or dreaming.
As opposed to a non immersive game which is closer to a film.
And in a dream I have minimal body awareness.
It's a waste of programing time to my mind.
But Bakerman, I'm willing to be proved wrong.
Ostriig on 17/9/2009 at 19:10
Quote Posted by ZylonBane
Try actually reading the paragraph you almost quoted. It answers all your questions.
No, it doesn't. You're giving two examples of affected gameplay - the ladders and the sidesteps, the latter of which I assume, as I haven't played DS for a while, happened when turning on a fixed point. The sidesteps while turning issue I've already pointed out could be easily addressed by an animation hierarchy related to the yaw from the direction you last moved in. The point is to fit the animations to the possible poses which stem from your controls, not the other way around, and if someone did manage to fuck that up, it's a problem with the implementation, not the concept.
Since Mirror's Edge got mentioned earlier, let's take a look at that, too. In what way did the implementation of body awareness hinder its gameplay or your immersion?
Quote:
Now here's a question for you: In the good Thief games, when on a ladder you can move all around it and look in any direction. How would that be animated with a body awareness model?
I haven't played T1/2, describe what the movement was and how it was generated from the controls, and I'll give it a shot.
Edit: And if you mean just something like it was in DX, as in roughly the same degrees of freedom, that's not hard to emulate.
New Horizon on 17/9/2009 at 19:35
Quote Posted by Ostriig
Since Mirror's Edge got mentioned earlier, let's take a look at that, too. In what way did the implementation of body awareness hinder its gameplay or your immersion?
Thief and Mirror's edge are two completely different styles of games. The point is, with any body 'awareness' setup, the player model has to be 'pulled into position' for the animation to line up with a lot of the physical world...if they even bother to do that. They did it for ladders in TDS, but not doors...so why bother having it in the first place if it doesn't work for everything? Because if you did, you would then have to deal with being too close to doors to open them and the player model banging into them or getting stuck against them. The developers would have to compensate by putting the player on rails while the model got put into position, opened the door, sidestepped and yadda yadda. For a game like Thief, you don't take control away from the player like that. that's why body awareness doesn't work for Thief. Plain old, traditional first person is perfect for the amount of control needed in a Thief styled game.
Our work on TDM is a fine example of just how little body awareness is needed if the gameplay is actually good.
Body awareness is a gimmick, and it is not needed. If the game is more cinematic, calling for quick time events and player on rails animated moments...then sure, it's a perfect thing to have...just not in a game requiring the type of control Thief does.
Quote:
I haven't played T1/2, describe what the movement was and how it was generated from the controls, and I'll give it a shot.
:laff: Epic fail!
Try it for yourself and find out.
