Ostriig on 19/9/2009 at 17:02
Quote Posted by Subjective Effect
I think it would actually be incredibly difficult to implement well actually. There are so many situations you'd have to code for - stairs, slopes, uneven ground, carpet, grass, small objects on the ground, standing on the edge of something and shuffling a little or turning around, climbing up ladders, sliding down ladders forwards or backwards.
It's be a mess or doing it right would cost too much time.
I don't think that's the case, many of those are a lot easier to solve than they seem. Stairs, slopes, uneven ground and small objects all get covered by the IK. For turning on a fixed point all you need is a turn-on-your-heels animation, and for turning while walking you need not bother altogether. And some of the more general animation assets could be shared with NPCs anyway.
Sure, there's more animation work to do, but I don't see it as a colossal undertaking to require actual compromises in other development areas.
Quote:
I'm also a fan of minimal HUDs but the lightgem is one thing I like to have. Have you seen the TDM HUD? That's nice and minimal.
Yeah, looks great. Nice and unintrusive, like a proper PC HUD should look like. Something that Bethesda have yet to wrap their heads around. But anyway, if I could, I'd probably even give it a shot at disabling
that. By the time a comprehensive, modular UI mod came out for Oblivion, I was comfy enough with the game that I decided to use it to disable pretty much
everything. All I had left was the title text popping up for objects you're targetting, and my status bars would each fade in if they dropped below 40%. But yeah, that's just personal preference, I'm not making a point here.
SubJeff on 19/9/2009 at 18:19
What the hell is "IK"?
I like the idea of things only popping up when you need to know, but it really depends on which game it is. In Metroid Prime, for example, or Deus Ex, the HUD is part of what makes the game. Especially in Metroid because it supposed to be like you're in the suit and its pretty restricting but the payoff is it gives you great power.
Ostriig on 19/9/2009 at 18:56
(
http://en.wikipedia.org/wiki/Inverse_kinematics) Inverse Kinematics, a process for determining the positioning of joints within an animated hierarchical structure (i.e. the animation skeleton). In this case, the application would allow for the feet to be accurately placed on a terrain with varying height levels, and the position of the knee joint would be dynamically determined to bend the leg naturally for that situation.
Here's a quick demo of how it works in a program called Poser. IK is enabled on the right arm, and notice that I've selected the hand (wrist node), it's the only one I'm manually placing. Upper row is start position, lower is end. The elbow joint is then automatically shifted in accordance to hierarchy (shoulder node is parent, it can't move it), bend rules (natural anatomy is enforced on the model), and the position of the hand node (lower in hierarchy, and the one I've moved). This is entirely real-time, by the way, it's not resource taxing.
Inline Image:
http://i211.photobucket.com/albums/bb67/Ostriig/poser_IK.jpgEdit:About the HUD thing, I think the point is for it to make sense. You could easily construct a consistent backstory for a corneal implant or contact lens, or the like, in various science fiction settings. Or various displays, like, for instance I remember being pleasantly surprised by an old UT beta, with every weapon having the ammo counter on it. But it really varies with the setting, it's harder to do in non-tech ones. The only indication as to your HP in Mirror's Edge was the degree of screen desaturation, it was superb, but it only worked there because quick health regen made sense in the runner gameplay, and you'd notice it. You put it in a game without regen, and within two minutes of staring at a slightly desaturated screen you'll have no idea what your health's like.
As far as Thief goes, though, I don't see many options to integrate the UI in the game world. A classic one is probably the way to go.
Unless you were to display a glimpse of the character's arm or hand at all times on-screen, to have something like a bracelet displayed. A gem on it, for obvious reasons, and a discretely varying number of links or studs could indicate HP, maybe arrows as well. But it might be more of a hassle than it's worth, with dubious results.
SubJeff on 19/9/2009 at 20:03
Well if you can implement that in the game and it works well then there is no reason not to have feet.
But only feet, and only for the express purpose of position on edges. And it should be smart and adjust for things like walking along a curb - no floating, no tripping, seamless walking normally without it interfering with headbob. There are some things we don't mind in RL that would be very, very irritating in game.
jtr7 on 19/9/2009 at 21:51
As long as the elbows and knees don't move inward past the hands and feet at the wrong time and poses aren't defying gravity, or bump into the environment, leading the devs to implement auto-positioning, which I don't see going away yet. A do not want to play a marionette and still do not want to play a robot. There are real world reasons these realisms are still not in games (or they are in rare cases but those games don't deliver on other aspects), and you know we aren't the only ones thinking about it. Money and the license purchasing of third-party applications to achieve this are what's necessary at this time. It's not important enough and puts more stuff in front of the camera to block the world out. This begs for a revisit to the PowerGlove controllers.
The only reason for a light gem and other HUD components to be large and intrusive are for players playing on television sets...and to look "awesome!"
SubJeff on 19/9/2009 at 22:28
I don't know about that (the HUD line).
Last gen maybe, but this gen consoles are often played on HD TVs and the HUDs don't need to be so in your face.
Just look at Assassin's Creed - even on a big CRT (I have a 32") the HUD is so tiny as to be unreadable. This is why I haven't played it yet (waiting til I get a HD LCD TV).
Ostriig on 19/9/2009 at 23:18
Quote Posted by jtr7
As long as the elbows and knees don't move inward past the hands and feet at the wrong time
Not the case, as specified in my previous post, containing a working example: (
http://www.ttlg.com/forums/showthread.php?p=1908804#post1908804) "bend rules (natural anatomy is enforced on the model)". Furthermore, we're talking about relatively small height variation, the sort you'd realistically be able to step onto.
Quote:
and poses aren't defying gravity,
If you're referring to feet hanging in mid-air at great height, a quick and comfy solution is to place the heel of the suspended foot behind that of the foot still on ground, at a 60' difference in orientation. And I imagine there are far more elegant ones, too.
Quote:
or bump into the environment,
Already addressed
many times, no changes to collision are incurred.
Quote:
leading the devs to implement auto-positioning,
No need. See Mirror's Edge, for instance - only used in a specific type of situation that is not part of Thief's mechanics, and on ladders, iirc, because the gameplay had no need of anything else. The latter can be implemented properly.
Quote:
Money and the license purchasing of third-party applications to achieve this are what's necessary at this time.
IK is a concept, not a patented or trademarked property, there's no licensing involved. And it's just the
better solution, standard preset animations are another. The relevant animation and coding resources are already required for development.
Quote:
It's not important enough
Nevermind subjectivity, to say it's not important
enough assumes you're weighing a pretty good idea on what implementation requires against the outcome, whereas here you seem to be just grabbing at random at any possibility that there might just be some horrible hurdle in the way. Which is really not the case. If you just say that you don't care about or for it, that's your personal preference, and I think everyone can respect that fact.
Quote:
and puts more stuff in front of the camera to block the world out.
(
http://s211.photobucket.com/albums/bb67/Ostriig/?action=view¤t=me_degree1.jpg) Barely the (
http://s211.photobucket.com/albums/bb67/Ostriig/?action=view¤t=me_crouch1.jpg) case. And with realistically no impact on gameplay. It's your
feet.
The reason you haven't seen this stuff before is because there was no call for it - what good would realistically reactive visible legs do when your head looked like it was chiseled out of a brick? Why invest the dev effort into it when "lip-sync" meant your character's mouth would flap open and closed like a duck's bill? Why put one or two extra skeleton joints in the foot, when the mesh that went with it was blocky and jagged anyway? Technical considerations meant that there was little possibility and/or point for the feature in the past. But that again there was a time when dynamic shadows started coming into common use, and rounded goblet meshes too.
Now things are different. The tech is here, the hardware support, and games have the visual fidelity to make such a feature a natural and reasonable addition.
Namdrol on 19/9/2009 at 23:55
People on both sides seem to be listening.
I could quote from the two differing opinions to show how things are approaching a common ground but it would take things out of the context of the ebb and flow of conversation.
But I want to try a far more common example of the limitations an increased body awareness would cause than a candle stick through bars. (Reasonable example though it was imho.)
Pictures in a frame. Garret has robbed these all through the series.
Does he get out his dagger out, cut round the frame and roll up the picture?
Look, I haven't been sitting trying to wrack my brains for the hardest thing I can think of.
It's just that as I see it, approaching a form of "on-screen" realism means that things like this become harder to implement, and therefore start getting left out.
And yes, we could sit and watch Garret do it (and drop out of immersion like a stone from the sky).
Or line up the knife and cut and roll it up ourselves whilst keeping an eye out for the guards.
But how much more complicated does each layer of action and movement and the minutiae of body placement become?
And at the expense of what?
I agree that times have moved on and if changes are made that mean we catch a glimpse of a body, if done well, of course they should be implemented.
But not at the expense of limiting the flow and immersion.
(And I still think my point about encumbrance is valid.
With our visible, well animated bodies how much can we carry? Where does realism stop?)
One of the best threads going at the moment to my mind.
It's forced me to think, and I hope learn.
SubJeff on 20/9/2009 at 00:04
Quote Posted by Namdrol
People on
both sides seem to be listening.
I could quote from the two differing opinions to show how things are approaching a common ground but it would take things out of the context of the ebb and flow of conversation.
Jesus Christ people! I think we just found the number 1 candidate for Official Thread Flow Commentator Extraordinaire.
Well done Namdol, you've successfully read the thread.
If the mods don't sign you up I'll have you. You can come into threads I'm posting in and occasionally remind us (and me) exactly what direction the thread is taking.
Its going to be awesome.
:p
come on have a sense of humour you deserved that
SubJeff on 20/9/2009 at 00:07
Back OT.
I forgot about Mirrors Edge since I only played the demo on the PS3. The feet in that were pretty unobtrusive so yeah, Ostriig, if done like that it's be fine.
Problem is - you just KNOW it wont be.
Just frakking know it.