One of the main challenge to tackle to achieve a dramatically interesting game is to extend the classical object economy found in games. Here is a simplified diagram that represents such an object economy:
 |
| Click to enlarge |
Here is what I will try to implement for my DramaticGames experiment:
 |
| Click to enlarge |
The huge difference with economies found in current games is the fact that this one is generalized; this means that the DramaManager AI can manipulate it as it wants to enhance the player experience.
Another difference is the fact that relationships and the way to change them are multiple. So, the goal of the game is not finding how to get the good actuator but: what actuators will be the best ones and why?
This economy is just one part of the equation. The other part is the “thinking” villain, the dynamic opposition that will transform the experience from being a puzzle to being a competition with the game antagonist…
What's your opinion on game like Tokimeki memorial who have a prototype of this gameplay?
RépondreSupprimerAlso game like sakura wars, valkyria chronicles, princess maker and many other. It would be cool if you give context to your analysis with previous implementation of those idea.
Unfortunately, modern game economy are barely more sophisticate that your locks and keys schematics.
Talking of context, I can't remember game where the player can directly affect the relationship between two npc (unless in script branching quest like in zelda majora's mask) outside the social experiment of Chris crowford (gossip).
RépondreSupprimerAnd the main question remain: what would be concrete gameplay about modifying relationship between NPC? I have a pending experiment on this i'm waiting to do (zazie la zizanie where the goal is to effect a goal relationship by abusing a social network without being taken).
Hi Timmy,
RépondreSupprimerThanks for the reference games you gave me. I will try to get information about them.
About the player affecting the relationships between NPCs, think of it as "gravity" forming solar systems. Those systems are “factions” that gravitate around a leader, the player or the NPC antagonist. Those factions are growing and shrinking depending of the attractively (actions) of their leaders. There are off course multi dimensional systems.
About the concrete gameplay, this is what i am designing right now but i can already tell you that it will works with macro RPS graphs representing the competition between the player and the antagonist. I am still using a murder case as a study example. I will post my findings as soon as when i will have something concrete...
Oh that's totally different of what I think you were doing :)
RépondreSupprimerI thought you where doing a simulated social landscape where the player must navigate by abusing the social network as a systematic key.
It's more like a diplomatic game with a war between two agent and many neutral social resource agent that "fight" along the two. Pretty much like the game we were trying to do with the mtv license.
Base on this it's more like the social system of baldur's gate 2 where social agent came, comment and left based on your reputation, your ally and action.
I will give you some information on those game to cut you the time to research them and filter the interesting bit. So you can focus on your model.
RépondreSupprimerTokimeki memorial is dating sims game, basically the goal of the game is to "get a girl" before the end of the school where one of them should declare herself to you. The game work by having the player manage his real life, a girl he want to seduce and the social network of girl around him. Ignoring friend of the target may piss the target but also spending too much time with them might piss her too, but if a girl is jealous sh might also modify the relatioship with the target. To spice thing the game have planned and random events that modify relationship base on player's action during those events. The system is pretty simple, each girl are on a fixed array of relationship with each other that simulate gossiping, what it really do is to trigger behavior based on relationship after the array modifier, if the resulting data is above or lower certain level it trigger a behavior. Jealously for example mean that a particular relation is too high and the particular girl relation is too low, the game express it by a "bomb" near the girl and all girl start to act distant to you, you must solve the bomb before having a good relation with every girl again (but it depend on who relate to who). It's a simple system but it work enough to give the illusion (the player is not aware of those background calculus).
Sakura wars is a bit simpler, but work in the same line, you can have buff with character by having high relationship to them, but you can't have high relationship with everyone as some relation hurt other. It's more about managing a faction efficiency.
Valkyria chronicle does not have much of player interaction beside order but does something interesting by having all troop nearly have some high stake background and stat that rely on that background to describe their efficiency and relationship. It's a wwar game and unit have explicit named relation (traits) to each other, these relation affect the buff/penalty a character can have (for exemple after seeing a friend dying) and his consistant with the personality of and background of a character. I'm not sure but there is also character quest that make a character evolve by changing acquiring new traits. Those traits are direct behavior in battle and does a great jobs at building the character and showing his personnality, having the player take them into account makes the player care about them because they are tied to their definition. Mechanically it's interesting because it's not a personnality/relationship simulated dichotomy, it's a direct behavior, each traits is self contained and really describe the character instead of having personality stats. Usually relationship model use an implicit model that let a system read the npc stat and decide how he react accordingly in front of another set of stat, valkyria is explicit, it say how a character react to a particular event that relate to a particular character, it only work because character are unique and the setting and context tightly controled and framed.
Those game have very simple mean to acheive something that let the player filling most of the blank (what happen outside his presence) by having a simple logic consistency. For exemple, we never see the girl chatting in tokimeki, that's something girl do so the array is a crude simulation that work pretty well. The presentation also goes a good way to mask the space time implausibility of those action (where does they met? how long, etc...) it makes those implied things acceptable because the framing of action is consistent with the game boundaries, it's the equivalent of ellipse in movie (we don't follow an addventurer crossing the desert, we have editing cut) that most story game fall into, except we also have an ellipse of space (but it could be argue that moving handle abstraction of space through editing and framing already).
Oh a side note on Tokimeki: The game work because all action are ambiguously social and framed with the romantic subject, it is also ambiguous because each girl have different threshold and relationship with each other, so there is a lot of psychological reading as the feedback is equally ambiguous. The player must assess the social context and the psychological context of his action to be successful in the game. Action also change meaning depend on the relationship level with a character.
RépondreSupprimerFor exemple, after each school day a girl might randomly pop and ask the player to walk with her, does this mean friendship or romantic interest, turning down a girl to walk with her might be rude if it's a great friend already but you don't want it to evolve into romantic territory (you can't get all all girl they gossip!) also some girl might only want to piss another girl because you like her too much. Also when a girl bomb or frown on you, did you make something to her or with one of her friend (over dating or ignoring)? All the guess keep the player on toes. and truly express the delicacy of social gameplay
The limitation is what make the game interesting, it transform the game into social system and psychological puzzle.
Thanks Timmy for all information.
RépondreSupprimerBTW: you speak japanese to understand games like Tokimeki ?
About Factions, the word i choosed is probably too strong. Let's say that it's about group dynamic. Groups can sometime be loosly formed and very volatile. In other context they can be very solid. It all depend...
Nop I don't speak Japanese, I played the gbc version and it was enough to get the game only through behavior and icons in the GUI! It show the mechanics was strong enough to overcome the language barrier.
RépondreSupprimerI forgot an important point in my behavior economy diagram: a relationship can inhibit an NPC behavior. For instance, if an NPC A like an NPC B more than he like the players, he will not do a given behavior to not “harm” NPC B, even if the player ask him.
RépondreSupprimerI can totally see what kind of game you want to do, it's all about social influence. But under the carpet it's still the same hard simulation.
RépondreSupprimerYou may have look there:
http://www.jorisdormans.nl/article.php?ref=machinations
http://www.jorisdormans.nl/machinations/wiki/index.php?title=Main_Page
You can totally simulate game mechanics and social dynamics out of those 5 game grammar element (sink, source, pool, flow, info) :)
I can see it's all down to the "level design" too, what's the main interaction and how you move field potential to create the desired events! For example modifying a field (character's disposition) to create desired behavior. Now it depend of the nature of the final goal and what or who host it (because it hold the value of action >>> adverb). What's the current goal of the setting? desire/fear to obtain/lose what? Is the murder case simply about finding who? something like that? http://www.squidi.net/three/entry.php?id=69
Basically it look like sympathy is the main key to unlock action (behavior) and maybe another factor (mood, equivalent to some buff debuff). The interface will be action to raise/decrease influence (generally sympathy but also fear, seduction or any persuasion mean toward a character) toward self or other, which allow access to data and or indirect action, and maybe ripple action toward a desire effects. I Can totally see the gameplay unfolding before my eyes, less like room and more like a dynamic circuit board. The gameplay loop would be gain data to gain or modify influence (for example exposing lie to change relationship, phoenix wright like) until the target action can be reach. I had work on this kind of system multiple time.
Basically the drama in your system arise from the identification of the player to the imply social setting.
Hi Timmy,
RépondreSupprimerThanks for the thought and the great links. There are a lot to read, so it will take me time :-)
I will definitely have to try the Machination Grammar (and/or the Stéphane Grammar).
I started to represent the gameplay as an RPS graph. I will try to post something soon. That’s said; maybe I will switch or complement it with the Machination/Stéphane Grammar.
Some important things to understand about my approach, which make it quite different from other game economies and other interactive story telling attempts, are:
• The opposition is dynamic. It’s not a puzzle. The antagonist thinks and also dynamically changes the social network. Offcourse, the challenge is to render this very understandable for the player.
• The game economy, the RPS and the thinking antagonist, are “just” the story engine. This is not where the drama takes place. The final goal is still: meaningful theme exploration and emotion generation. Those will takes place in other story threads implemented by NPC characters that has specific roles for the Drama. I am still thinking of using the Dramatica model for that. This will be off course merged with the story engine. As in non-interactive stories, where the different story threads are interleaved and mutually dependent.
• I am still planning to use the Quantum Uncertainty metaphor on this game economy to regulate the experience pace and to insure player centricity (if things doesn’t happen under the player eyes, they are too difficult for him to understand). I successfully implemented a prototype of Quantum Uncertainty for a key/doors economy and the result are very promising: the keys/doors layout is always different at each play and is always “interesting” (as much as a key/door mechanic can be; lol). But using the same principle for a behavior economy is quite another story ;-) Indeed, the elements with uncertainty have more links in-between them. We will see how it turns out…
The machination grammar is a refinement of the stephane grammar, the author studied the shortcoming and address them to generalize them. Stephane is in the credit!
RépondreSupprimerI'm aware of the dynamic aspect, it will have the same problem as a RTS but with ambiguity on unit (which one is on my side. You either need to make all transformation public (like in a court), create hierarchical role and important (some character are proxy, and it will create focus which is necessary for drama, no story have more than 5 at most important character or they are distribute in arc and story thread), or expose an abstract interface that track the relationship. Or the game should be like tokimeki a game of tell and guess, basically the player must pay attention to ambiguous data that hint shift in stance and must spend resource on probing "truth", The balancing between certainty and doubt is what keep the player on toe. Basically it mean cross checking fact and behavior based on context and perceive agency. Also take a look at the series "lie to me".
Tips: adapt and merge the dramatica model with the greimas model ;) it will unleash further clarity. Look at the section additional story point to do it (goal, consequence, cost, dividend, requirement, prerequisite, precondition, forewarnings)
Sooner or later you will find game and story have the same structure and one only enrich the other. It will be way simpler to distribute fct and create drama.
When I say goal, I meant the overall story goal around the murder, a backbone is still important, even if the game is about exploration of emotion, you still need a hard cold structure to support it. Laying that structure help to focus only on the interesting part, even if you modify that structure to better accommodate your need after.
Quantum theory or you may name it author hand, many serie use this, lost is a big big guilty contender, where people witness something important and does nothing until a climax is required.
Dear professor Neoshaman,
RépondreSupprimerIt's been a long time since i noticed that game structure and story structure could be partially unified ;-)
For the rest of your comments, there are a lot of very valid points.
My next goal is to finish this damn story engine diagram. This should reply to most of your points on the number of characters and so on...
Regards,
-- Your beloved student.
Oups I don't want to come out as condescending, sorry for that. As long you appreciate my point as valid I hope yu will forgive me for that.
RépondreSupprimerBut we still diverge on story vs game, I claim there are not partially but the same structure.