The Oz papers tell me that expressing emotions is very important for the characters to be believable (which is something they in turn learned from the Disney animators). And my Body Language book tells me that when people like each other, they start imitating each other’s body language, at least in the early stage of the friendship.
So it seems that it would help believabilty if characters would express the effect of an interaction with each other through body language (rather than by having plus or minus signs float above their heads as in The Sims). So that when the interaction (i.e. “achieving the goal” in Consumerism speak) results in a positive effect on the relationship, the characters briefly imitate each other’s posture. Or when the effect was negative, they purposely don’t.
Perhaps the animation sequence that is played during such an interaction can end in a certain pose (it has to go back to some neutral stance anyway) which is an invitation for the other character to either imitate or ignore.
Even if this would look artificial, it could still be powerful. I only wonder whether this can be read by the player (either consciously or subconsciously). I know of only one other game that uses imitation to express relationship: in Animal Crossing a character changes into the other character’s clothing to express fondess. This is very clear to the player.
Posted on June 13, 2006 at 9:00 am
There’s a problem with implementing this in the “Consumerism”, though. In that model, a relationship is described in a more “abstract” way. It is in fact just an aid for the system to guide consistency in the behaviour of the characters and not a real relationship.
One of the things that happens, e.g., is that when both characters agree to do something bad to each other because they hate each other, the effect on the relationship would be positive. On a deeper psychological level this sort of makes sense, but instinctively we consider only nice things to be good for a relationship. We don’t consider being enemies a good relationship, even not when both parties agree to be enemies.
The solution could be simple: only run the imitation routine for characters that have a “good” relationship with each other. And only after a positive interaction. All the other interactions will end in different final poses automatically anyway.
Posted on June 13, 2006 at 11:43 am
I think we should definitely run a few tests with characters imitating each other and the player. This could be the animation they display, or maybe even the ‘opportunity’ they radiate, affecting other characters.
And we should also run a few tests in which certain characters do exactly the ‘opposite’ of the one they ‘imitate’. I think these opposing characters will make for very interesting relations. Players will probably like antagonists quite a bit, I think they will seem very autonomous and convincing.
Posted on June 13, 2006 at 12:10 pm
I’m not sure who you’re referring to as “we” in your comment. I welcome your suggestions but Auriea and I need to control the design of this project completely if we are to actually make something.
So far, I’m thinking of the player’s avatar as another Drama Princess with the only difference that the player takes part in the decision process. But the avatar will also have some autonomy in the way it responds to things e.g.
The imitation I was talking about was quite literally playing the same animation. What do you mean by “radiating opportunity”?
In Tale of Tales we always prefer harmony over antagonism (without fanaticism, though). But what exactly do you mean by “opposite” behaviour? Can you give an example?
Posted on June 13, 2006 at 12:39 pm
I apologise for suggesting that I would be part of the design team. It was probably a slip of the ‘tongue’ because I’m excited about the project. =)
I know that one of the main goals or principles of Drama Princess is, instead of creating AI, to create believable character behaviour (mostly in the form of animations).
About what I said above: I would imagine that on a very high level a character would still have a state, which could influence other characters.
So a running character would not only influence another character to run, it would also make the other character influence other characters to run.
Antagonism will, to me, seem like ‘conscious randomness’. It will also break up the harmony, something which I myself am not particularly a fan of. Just like in art, using a fixed palette is nice, but it can be accentuated by using a complementary color.
Example:
Character A is happy and dancing. Dancing past character B, A influences B to start dancing as well. This is imitation.
When A dances past C, C responds by standing still, or by becoming grumpy, or by imitating A with a purposely bad dance. There is definitely a relationship between both A and B on the one hand, and A and C on the other. I would be impressed by both, and even more so by the one between A and C.
Another thing to consider is to give the player a sense of belonging, or of being unique (purpose?), by being imitated, or by having his avatar imitate the general mood / animation of the characters around him.
Example:
By having the other characters play a depressed mood, and let the avatar walk around proudly, the player might get a sense of heroism, of hope. He might be more inclined to proactively continue his journey in the game world.
Posted on June 13, 2006 at 12:52 pm
To expand on the topic of harmony and antagonism:
Instead of just copy or oppose a scale could be possible. In an abstract sense, a character could use a random function to decide whether it will imitate, oppose or ignore another character.
However, I’m always a fan of keeping things simple, and controllable. I think that most characters should have a tendency to choose one of the options, and not randomise on a scale from 0 to 1.
Looking from another point of view – I think it would be perfectly valid to have categories of ‘influences’. A happy dancer might be ignored, but someone running for his life might not.
Posted on June 13, 2006 at 1:29 pm
Actually, in our current thinking, a character wouldn’t really have a state, apart, perhaps from being in the state of doing something with an object or with a character. The only consistent and growing things would be the relationships it has with each other element in the world. Interactions with these elements would alter those relationship. If the element was a character and the interaction ended in a positive alteration of the relationship, both characters could play the same animation after the interaction, to express their fondness of each other. This is the only time when I would use imitation.
Although I like your idea of contageous dancing. 🙂
I understand what you mean by the example. And I think this would already be possible in the current model. But it wouldn’t be part of the imitation principle. It’s a matter of content. Character A would be advertising all sorts of opportunities (I guess which ones will depend on some sort of state), including mocking his dancing. If character C doesn’t like character A, there’s a good chance it chooses that opportunity as a goal.
So relationships are the center of all decision making in Drama Princess, as well as the only place where change happens. And each relationship is expressed by a single number (from -1 for enemies to 1 or more for friends). I even think that this number should be the same for both members of the relationship. But it could possibly be different as well to allow for assymetrical relationships.
So whether or not “a happy dancer” would be ignored by a character would depend on the relationship that this character has with the happy dancer.
Does this make sense?
Posted on June 13, 2006 at 2:08 pm
That makes perfect sense, and the idea of defining relationships rather than character states sounds really exciting!
I can’t wait to build and watch a test in Quest3D.
Did you get my mail about visiting you?
Posted on June 13, 2006 at 2:41 pm
I think I’ll try to build a text-based version of the decision engine first, in Javascript or something. Just to see if that part works and to finetune it. But first the design needs to be finalized. There’s still a few weak spots. I’ll try and post a concise version of the latest model later this week.
Yes, I did get your mail. I’ll answer later today.
Posted on June 13, 2006 at 5:54 pm
I’m planning on using 2D sprites for Magic Circle, so transition animations aren’t so much of an issue. The postures will also be digital, instead of analogue, which is much more nuanced but much more difficult to integrate. There’ll be 9, four orthogonal ones, four hybrids, and one nuetral posture. The imitation you’re talking about would apply to early friendship, petitions for help, team co-ordination ect. Its a great dimensions of feedback for the players. Relative distance is another, how far people stand from each other says a lot about their relationship.
Posted on June 13, 2006 at 5:57 pm
You really need to tell us more about this Magic Circle… 🙂
Posted on June 13, 2006 at 11:46 pm
Well heres a post on the upcoming engine prototype: http://projectperko.blogspot.com/2006/06/rocket-heart.html
I can’t claim to be the genuis to make the core of it happen. I consider myself lucky to be able to bring a UI design and characters to the table.