Platinumoxicity on 18/9/2009 at 15:35

---

Thief 4 can have "body awareness" but only if it has been designed with "Mirror's Edge" like quality. TDS body awareness was designed in more like "ET on Atari 2600" quality. By all means, waste time and money making a body awareness system, but let it be included in the final game only if it's better or at least as good as basic 1st person mode in every single aspect imaginable.

What they did with TDS body awareness was basically that they took a perfectly working car, took out the round wheels and replaced them with irregularly shaped wheels of different sizes and made the car 3 times as high as it used to be. Looking at someone driving one sure looks funny from a distance, but the driver sure can't get anywhere with comfort.

Ostriig on 18/9/2009 at 16:57

---

Which is what I've been arguing for from the start. If you're not fine with calling that "body awareness", then fine, call it "visible body" or whatever you like.


(http://www.mirrorsedge.com/) And still. Entirely different games, I know you've already pointed that out and I agree, but for the purpose of positioning and seeing your legs and torso, it works. Maybe Doom 3 doesn't have the proper animations framework, or maybe you didn't have the tools or know-how to properly implement this feature. This is not meant as an insult, by the way, it's just a possibility I'm asking you to consider. I take it back that you don't know anything about animations, but the point still stands that it's not necessary to lock the player in while one's running.

Also note here that what I'm suggesting would also require more than just traditionally recorded animations. Typically, whether it's the recorded positions of the mesh vertices or the recorded positions of joints in a hierarchical skeleton, the only variable they play relative to is the player position. Some of the stuff I'm talking about would require runtime calculations of joint positions relative to both player location and the position of the object you're interacting with. Have a look at the door example I've posted for Wormrat some paragraphs below.


I've seen T2 played at Bikerdude's and I can't say anything about the control and movement jumped at me as being so different from, say, DX. But, if you want to show me otherwise, give me some examples. Describe some situations where Thief is different and could not employ a visible, animated player body, a situation where the inclusion of such a feature would prevent you from preforming an action you'd realistically expect to. Because, really, so far it's only Wormrat that has given some situational arguments to discuss.


No, nothing like that. If you have small, immovable objects, then your feet would not dance around them, but step on them. It's an application of IK starting from the fixed point that is the height level below the character's foot, continuing along the movable joint that is the knee, and ending in the other fixed point that is your character's hip node. Thus, your body, collision box, and viewpoint would remain consistent with the actual walkmesh below it, even though your feet would accurately be placed over any small volumes on the ground.

Edit: I take that back about Fallout 3, though. I took a closer look, and it does raise the character's feet to step on stuff that's below it, but it still uses preset, recorded joint positions, rather than dynamically generate them at runtime. It's still perfectly doable, but F3 doesn't have it.

Now, if your ground objects are movable, the standard collision box will knock them over anyway, regardless of whether you have visible legs or not.


No, hold on a sec. I've mentioned this before - I'm not arguing for any changes to control or collision. Your arms, if included, would not have any additional collision. Remember that your already existing bounding model is constructed to mimic the proportions of a player body anyhow.

Again, it's what ZylonBane has rephrased a few posts up - the "visible body model should just be "along for the ride", and not be allowed a single shred of influence over player movement".


There could be situations where an animation would have to be interrupted, but I expect they would be rare. You could solve them by either having an "aborted" animation to play, or by letting the animation play with a generic fixed point that would fit it.

Let's consider the door opening example. The start point is the door knob, the end point is the character's shoulder, and we have two segments in between, arm and forearm, connected by the elbow joint. Now, in order to open the door, with regards to animation, you have to turn the doorknob and then give it a slight push. The prerequisites are that you are in range. The calculations behind the animation are: place the hand object on the doorknob, which puts the wrist joint (start point) in position, then calculate the required position of the elbow joint as such to accommodate the restrictions of the elbow's bend capabilities and the given position of the shoulder (the end point, which is given by the character's position). Then, simply play a rotation of the hand element, a slight push, and return the arm off-screen. It sounds more complicated than it is.

Now, what if the player pulls back right in that, say, one second that the animation is playing? Well, if he pulls back a little, and is still in range, all that needs be adjusted is the elbow position. If he pulls back a lot, out of range, then, admittedly, you'd either have cut to the return animation and let the door open on its own (having been activated), or play the animation with the last start point (wrist/knob) recorded that was still in range. Same as if you were bunny-hopping around the door. But how often would/should that happen?

Bottom line, you'd probably be able to do anything you'd realistically expect to be able to, and situations where the gameplay could not be accurately served by the body animations would be rare.


That's where I disagree, I just don't think it's that farfetched. I think it's well within the reach of devs today, and doable within the time and financial constraints that fit such an, admittedly, not first-priority feature.


I'll concede that point. In the event that your foot should obstruct exactly a (small) point of interest, it would require you to move off it. But I think that would likely be a very rare occurrence.

Ostriig on 18/9/2009 at 18:49

---

I'm sorry, I still don't follow you here. If the objects are immovable but have collision, you step on them. If they're movable, then they get kicked out of the way by your character's collision element anyhow, in current games with no visible feet, too. If the next generation of games will put more emphasis on stuff being knocked over by a moving player, then devs will need to design cleaner rooms or debris without effects, and that will be the case regardless of whether you can see your legs or not, because the collision box is present.


Yes, that's a good point there. I think it would be a fair amount of work to get the foot's that off the walkmesh to be positioned naturally back onto the walkmesh. For my part, though, even if this sort of thing wasn't accounted for, I'd still rather have the occasional situation where I've got a foot in mid-air rather than have no feet at all. But this is a matter of personal preference, and I wouldn't argue with you if you thought otherwise.

As for gameplay benefits, there's the one with knowing where you're standing, yes, and I mentioned earlier that another minor bonus might be using your body as the lightgem if you opted for a minimal HUD. Maybe I misspoke when I said "massive improvement", but between that and the added "realism", I still hold it's a worthwhile pursuit.


Before we go on, let me just state that when I popped into the thread declaring that "every game should have body awareness", I was referring to looking down and seeing the rest of yourself, I hadn't thought of visible arms when handling stuff. Then arms did get mentioned and I decided that's probably also doable, though I'll agree it's much harder altogether.


Now, yeah, if you wanted to be able to reach in through the bars with a visible arm and grab an object, that would take some pretty smart coding. And, yeah, I can see having to extract object from narrow spaces as a more common occurrence, so it would be a must. In fact, a general consideration, code-wise, would have to be given to other collision objects around the one you're trying to interact with.

However, if you're also referring to moving the candlestick inside the cage, I'll say that's not doable (not reasonably, anyway), but that I also don't think it should be. If I want to move the item from one end of the cage to another, why not just pull it out and then place it in the other end? That's why when I asked New Horizon for examples I mentioned they should be actions he'd "realistically" expect to be able to perform, and I don't think that a limitation of this sort would genuinely affect gameplay or level design.

jtr7 on 18/9/2009 at 23:48

---

Since moving the candle through the cage in real life would have most people using both hands, trading back and forth whenever one is more efficient than the other, I would gladly do without body awareness, and just have the candle guided through the cage. Freedom of movement, increased movement, smoother movement--thrown away for appearances??!

Increased environmental interaction with body awareness means more and more animations. Can't they see how much it will consume development and playtesting time? Unbelievable.:tsktsk:

The arguments for body awareness and 3rd-person are not realistic (except for how each individual perceives these is a real experience in their brain for them), do not account for real-world technical aspects, logistics, and ramifications. They are shallow and...dishonest. The desire to cripple a potentially great game for something superficial, difficult to implement, and test...is inexcusable.


I continue to pace about with my bullhorn, protesting the boring shallowness, and this hatred of more than half of what makes the games. Leave unimaginative realism out of the unique escapist fantasy. You guys do realize that adding more realism creates a desire for more realism, don't you? The closer it gets to real, things look more and more irritatingly incorrect. It will never be good enough. When you go to a play in a theatre, do you criticize the lack of realism? Or do you have an imagination that happily fills in the blanks and give in to the experience, immersing yourself, creating a powerful experience?

http://i70.photobucket.com/albums/i106/jtr7/OAF_Arrow.gif

It should concern us that we have among us those who cannot distinguish their feelings from the facts that exist without their input or observation. When I speak of Thief lore, I know I'm filling in the blanks, and when even the strictest canon is dismissed by them as irrelevant, they think their boring-ass views are superior. They can believe what they want for themselves, but complaining about more than half of what Thief is...is asking for backlash, and bewilderment of why these forums have this peanut gallery of anti-Thief hecklers as fans in thin disguise. I refrain from calling these guys trolls even though they share many traits, because they are also contributing to the community in ways I'm very grateful for.

Damn, this popcorn's stale.

The difference between the ridiculousness of my above pseudo-frothing rants and the counter-arguers, is that I know I'm being ridiculous in my delivery of the truth, and the arguers don't even know how ridiculous they are in their over-simplifications, and complaints regarding several fundamentals they wish to cut and replace with absolutely non-Thief elements--approaching a GTA style medieval game without the "A" for "Auto"--unless it's closer to "Automated." Handicapped by choice (giving them the benefit of the doubt), they continually act as if they really believe what they wish to see is simple to create and won't harm anything.


GO DARK MOD TEAM!!

Bakerman on 19/9/2009 at 01:12

---

Platinumoxicity - I basically agree.


How much was this due to leaning, mantling, and the slight feeling of weight that the Thief controls had? In DX, for example, I feel extremely rigid because all the character's actions happen so darn fast, including starting and stopping. Thief's physics were a little more sedate.


I would consider that a fair consequence of being a careless taffer and swinging your arms around like you're on a fairground ride. Now, I would be perfectly happy if the actual collision and control remained the same, as Ostriig is suggesting - but a visible body could be used to enhance your interaction with the environment, like in this example.


There are solutions - how 'bout if you're looking at your own foot, the engine realises this and the foot does a little step backwards or similar, so you can see that piece of floor you really have to see?

theBlackman - no worries, thanks!


I wouldn't even bother, because that character obviously has supernatural mind powers. I really love the degree of freedom you have there, rotating the object on all axes - and that's certainly possible to animate. But reaching extreme distances through bars that would barely accommodate an arm anyway is a little far-fetched.


It's more immersive? That was all I wanted to suggest :p


I would argue that body awareness is far and away more realistic. Because in real life, where things are absolutely realistic, I can see myself. QED. But I do appreciate the divide - gameplay vs. realism is something I'm struggling with a lot, and in this case realism won out - though I do believe you can have both if you do it right. We'll see how that goes :p.

jtr7 - we're on the right track, you're starting to make sense. Please continue to do so.

I just quoted that because there seems to be some sort of irony there that I can't be bothered figuring out.


Yes, increased movement would probably need to be thrown away in ridiculous cases like the TDM candlestick. I don't see that as a huge disadvantage, though. And who said anything abour free or smooth movement? Your personal misconceptions seem to be clouding your arguments. Body awareness does not need to affect the controls or smoothness of movement - of course, in some cases it absolutely should, but you seem to be of the opinion that any form of body awareness will automatically reduce your silky-smooth Garrett footsteps to jerky, clog-wearing square dances.


Or, like Ostriig said, they could use IK, which could actually reduce the number of animations. But I don't think the number of extra animations is too huge an amount for a professional studio. (Maybe if they mocapped them it would help, silly EM :p.)


Ahem. Please see the rest of the thread for details in this area.


Yes, but I trust that EM will know where to stop in that regard. And um... what's the problem with that? I desire for Crysis to have better graphics so it would actually look photorealistic, but that doesn't stop me enjoying its current graphical prowess.


I appreciate the acting, the music, where appropriate, and the efforts that the stage and costume designers have gont to to evoke the feeling of the setting within the constraints of the real world. Obviously I wouldn't expect a play to take me to 17th century Venice so I could be properly immersed - but I don't think that body awareness is quite as much of a demand as that.


Yes.


Very funny, captain. Nice try, but how does wanting a visible body translate to wanting Grand Thief Automated?


If I've ever said it would be simple, I apologise - I should have known better, since I'm strugging with it myself. But I don't believe this is simple, merely doable. I also don't believe that the impact it has on other aspects of development will ruin the game in any way.

Sorry if I sound condescending or sarcastic in replying, but I take offense at your rhetoric and general attitude. I can believe you're just being hyperbolic because you feel strongly about this, but I'd still appreciate being talked to like you talk to everyone else.