Read Build It to Understand It: Ludology Meets

[ETdoFresh]

Narratology in Game Design Space

[ETdoFresh]

https://eis.ucsc.edu/papers/MateasSternDIGRA05.pdf

[ETdoFresh]

Summary

Build It to Understand It: Ludology Meets Narratology in Game Design Space Notes

This paper is a look at how Facade adds yet another point in the game design space that is not fully defined or explored.

• Looking at existing games is limiting as it would be problematic to reason or theorize about potential games outside of this space.
• Cannot wait for AAA studios as they are mainly concerned with marketing and economic concerns.
• Designing experimental games can help explore these regions not yet explored by commercial games.
• Exterme Ludology Position - Games have nothing to do with Narrative
  • Argues narratives are simply a linear layer on top of agency ruch game play
• Facade aimed to explore non-linear, player-responsive narrative using a real-time, high agency interactive drama.
• Agency
  • Player Local Agency - Player sees immediate clear reactions to her interaction.
  • Player Global Agency - The long-term sequence of events experienced by player is strongly determined by player interaction.
• Strong Narratives
  • Complex sequences of cause and effect
  • Complex character relationships
  • Complex sequences of character interactions
  • Since player interaction can disrupt narrative at any moment, to maintain a narratives structure, it must restrict the player from affecting it
  • Hence removing global agency, Forcing player down predetermined paths
  • Hence narrative cannot be the heart of the game, but a layer above core gameplay action
• Shallow Games vs Deep Games [my terminology :p]
  • Games contain actions like running, jumping, shooting, etc
  • Most of these actions have clear effects, like changing the numeric scores or bars on the screen.
  • Social games may also use numeric scores, and maybe several layers of them...
  • But not communicated as numbers, but dramatic performances
• Facade is a two part game where
  • First part is a zero sum game where Grace and Trip are learning who's side you are on
  • Second part is a therapy game where Grace and Trip are realizing their own problems based on your actions
  • The whole time there's a hot-button game where player can learn about the characters and backstory
  • The game is also keeping track of tension level
  • However, if player gets too out of control, the story breaks and Grace and Trip throw the player out their apartment
• Facade uses
  • Multiple Fronts of Progression and Narrative Suqeuencers
  • Procedural and Keyframe animations of non-photorealistic rendering style
  • First person user UI using keyboard to speak nd navigate and mouse to gesture and use objects
  • Low Level Reactive Behaviors
  • Long Term Autonomous Behaviors
  • Joint Dialog Behaviors
  • Natural Language Processing
  • Beats
  • Mix-in Progression, mix in reactions to players actions [doesn't have to be immediate]
  • Discourse - Choosing best reaction to players actions
  • Drama Management - Regular choose from collection of beats to match tension arc
• Facade supports a high level of local agency at the expense of global agency [27 beats]
  • Hence a more generative system would be required in the future
  • It was challenging to communicate social state to the player [since it's not simple numbers or bars, like the sims]
• The Game Design Space
  • Each point in this space is a game and all of its specific design such as features and design decisions that make up the game.
  • It is not a 3d space, or a uniform space, hence not all the dimensions are known or at what scale they exist
  • Additionally, innovative games define areas that are not known which changes the dimensionality and scale of this space
• Game Design as a Wicked Problem
  • No Definitive Statement of a wicked problem: Create a game in which you're an unwitting guest at a couple's marital meltdown... OK?
  • No Stopping Rule... what is good enough? when do we stop?
  • Solutions are not correct/incorrect. Maybe better or worse?
  • Each wicked problem is unique, unless an absolute copy of a previous game [tame]
  • No Immediate or Ultimate Test of a Solution
  • Because of this, game design uses rule of thumb, case studies, and best practices in making the design
  • Experimental games play a significant role in mapping game design space
• Mission: Create an interactive story in which the player experiences both local and global agency
  • What are we trying to answer? How do we solve it? Maybe we throw definitions about games here?
  • Defn1 - Interactive Story is impossible [Lud]
  • Defn2 - All games are symbolic/potential narratives
  • Defn1 denies the nature of a wicked problem
  • Defn2 denies there is a problem to be solved
  • Perhaps we view what's out there already?
  • We find that most games are linear/semilinear stories layered on top of gameplay
  • But as mentioned earlier we are only looking at a small subset of potential game design space in this case
  • So our option to explore new regions is to make the actual game in the space
  • Design and Architecture
  • It is helpful to think of both of these dependant on each other
  • Design of any kind assumes an architecture. Even a paper and pencil prototype
  • Architecture provides a language for which the design occurs
  • An architecture is said not to support a design if designer is fighting or hacking his way to express his thoughts
  • Design and Architecture mutually constrain each other
  • Mapping Existing Design Regions
  • Playing games may not allow one to fully map local design space.
  • The design space is also emodied in the architecture

[ETdoFresh]

Thoughts

Random Thoughts during and after Reading

Communication of Social State

• How well is a game communicating social state?
• Should we run a study about that in InverseRPG?
• Comparing what the numbers/bars are vs what the player thinks the numbers/bars are for our HeroAI and other characters?

Generating Content

• One of the problems in Facade was generating the actual content leading to a less than desired global player agency
• How do we fix this? Lower level reusable actions?
• Progressive Generation [as a side note]
  • I was thinking maybe we allow the game designer/maker to author instantions progressively [bad example ahead :p]
    • Step 1. Author decides to make an animation for TALK [regardless of social state]
    • The author later decides there should be a different animation for TALK when the character has a high level of anger
    • Step 2. Allow the author to make a new animation for TALK when angry using social state, but the domain doesn't change
• Other PCG? Randomly placed NPCs? Randomly placed buildings? towns? caves? etc?

Design and Architecture

• Another point in this paper is talking about how design and architecture are mutually constraining
• We are using Unity and preferably (for me at least) WebGL as the platform to run on for InverseRPG
• This constrains how we design InverseRPG, but those limits are soo hard coded in my head, I'm really not too worried about continuing knowing this
  • There's still N dimensions of game design left, and I will proceed onwards!

Tension Curve

• I like this concept as it gives the story a structure, and informs drama/experience manager
• I think I like this idea as i think this assumes a "beginning, middle, and end"

The Abstract Graph Database [Big ol' tangent, skip if you want!]

• Imagine there was an abstracted graph database in the cloud that was forever expanding
• We could continuously add or remove hardware to expand the search
• The search happens forever, forever expanding the graphs it stores
• The goal: Create an API that gets optimal solutions using your own custom named elements in the graphs (ie states with transitioning actions/edges that have precondition and effects)
  • The API would allow you to download the whole graph [already computed]
• Heuristics would speed up the graph search algorithm, and if approved by a community, it will be the master algoritm performing the search
• Graph must support
  • multivalued variables
  • node pointers
• Numerics should be handled outside the database using limits and numeric graph compilers
• Would be helpful on generating plot graphs, finding solutions, etc

Lore Score [Another tangent]

• Not my idea, but bouncing ideas off a buddy, and he was saying we should track story lines via some Lore Score [or story progression score] to help make decisions.