Slashdot Mirror
How Perlin's Law Makes Gaming Credible
simoniker writes "Veteran game designer Ernest Adams has posted a new column on 'Perlin's Law' which suggests that all books, movies, and games have a 'credibility budget'. For games, both the designer and the player decide what happens: '...the story itself can only tolerate a certain amount of improbability before the credibility budget is exhausted, and the story is ruined.' According to this new law, named after Ken Perlin, who gave birth to the concept, games should not be infinitely wide-ranging or allow the player to do anything he wants."
I've always secretly hoped that games would one day evolve to a point of them becoming specific to the user. This "mind game" that Ender played had seemingly limitless possibilities and also seemed to reflect the user's psyche back at them and cause them to make connections they never knew existed.
Maybe the next step for video game engines isn't graphics rendering but instead, stimulus/response rendering? Where by the game reacts to user input using rules, heuristics and a bit of randomness and the game states are loosely defined. Why is Spore so popular? Possibly because of the number of proposed outcomes of the game.
We're no where near this kind of game play yet but it may be possible in the future. Perlin's Law seems kind of like a restriction that I honestly wish game developers and publishers wouldn't try to adhere to. Only when people take chances and think outside of the box will we find true gems in the video games. I'm sick of repackaged games and ideas.
My work here is dung.
The story might be ruined, but the public will still pay you money for a totally incredible story. Just look at the lasting popularity of X-Files, which drastically changed its plotline every couple of seasons (first greys, then black oil, then super-soldiers), and the current hype about Lost, which appears to be doing the same (first mystery island, then DHARMA, then Widmore, now according to producer podcasts it's soon to be previous inhabitants).
Really, is that any worse defining yourself by your favorite bands or TV shows?
The issue isn't the number of possible actions that your character can take. Those are good.
The issue is that those actions have only extremely limited and unrealistic results in the game world. What we need aren't restrictions on what the player can do (returning back to older games), but rather an improvement in how games react dynamically to unexpected user input.
Real life is not a state machine, moving from one state to another on linear paths. Games that try to be as expansive, or more so, than real life need to also not simply be state machines.
The ringing of the division bell has begun... -PF
Take the cheats from just about any game. How often does playing with cheats 'take the fun out of it', and how often does it 'improve the experience'. Given a cheat that lets you go anywhere and do anything in a game, it's been my experience that more likely than not, it ruins the experience (unless the game is obsenely hard, which is relatively rare). That also lends itself to a game balancing issue. The more open-ended you make a game (e.g., Oblivion/Morrowind), the more chances you have for both bugs, and balancing issues. (With enough grinding, it wasn't hard to become a virtual god in Morrowind.)
No matter how much you start out with you must never cross the line and have a character do something that is inconsistent with the world in the story. You cannot have a character from The Heart Is A Lonely Hunter leap into the air and fly. You MIGHT be able to get away with that in a Star Trek story.
We have always been at war with Eurasia!
This seems to conflate two very different ideas, one of which is obvious, the other seems misguided. Clearly, if there are no constraints at all on what players can do, that's going to strain credibility, in and of itself (at least, if complete freedom isn't limited to a special distinct mode designed for editing the environment rather than intended for "playing the game").
But I don't see how an increasing scope is countradindicated, so long as items in the game are designed for credible behavior and reaction. Sure, infinitely wide ranging requires infinite programming to create credible behavior, but its a nonsense limit anyway, since you'd need an infinitely powerful computer to run the game, and infinite media capacity to deliver it, anyhow. "You shouldn't do things that are impossible" isn't really a necessary warning.
. . .how do we balance the player's desire for freedom with the designer's desire to tell a consistent, coherent story. . . .
This tension is an essential element of classical (Freudian) phychology. Substitute the terms Id and Ego for player and designer, respectively. Indeed, in a post-structuralist view (informed by Lacan), any discourse structured as a narrative (that is, nearly all, internal or external), Perlin's Law offers interpretive value. For example, a measure of the bounds of normativity for an internal discourse (whether you consider yourself crazy) is a function of Perlin's Law over the constituent terms of that internal narrative.
Further study: Can we apply the concept to shared narratives like normative social behavior or political formation? Is the concept redundant with the contributions of the Frankfurt School?
Extra credit: Does this idea offer a description of the development of political reaction in response to sharply divergent, even orthogonal, shared narratives (q.v.--the Bush team vs. *the reality-based community*)? Is it persuasive?
Indeed, credibility has been a consistent focus of Rhetoric since the inception of the Western cultural tradition. Perhaps Mr. Perlin's own modesty should prevail over the enthusiasms of the geek community in general and Mr. Adams in particular?
illegitimii non ingravare
I think there is a certain amount of usefulness to this notion, however what I think it needs to clarify is that when the player is what causes the improbable action, it spends far less of the 'credibility budget' then when the game seems to be the impetus.
When I drive off a ramp, flip over and cause a 15 car explosion in GTA, it doesn't really affect my notion of the game as a vaguely believable caricature of America. However, if that happened all around me constantly it would bust that and I'd feel like I were in crazy stunt world or something.
I think that the difference in credibility effect between player impetus and game impetus is so great that the mere suggestion that player freedom is a bad thing is almost entirely busted.
It certainly makes it more difficult for the -game- to respond to the player in credible ways, but it isn't directly what the player did that hurts that credibility.
I'd say that arbitrarily limiting a player's freedom has a credibility damaging effect as well, since you feel like you are in an invisible straight jacket whenever there exists a mind numbingly obvious solution to a problem that can only be dealt with in the circuitous manner decided by the game developer.
I think you're being unfair- this article poses several tangible extensions to the "suspension of disbelief" concept:
1. Credibility can be treated as a quantifiable substance that can be codeified in a game
2. In interactive fiction, both the developer and player draw from a common pool of credibility, making it unique from other fiction
3. Players can destroy their own enjoyment of the game by using playing strategies that lead to wins but hurt the story telling element- Telling a story and beating a game are two separate ideas and interactive games struggle to accomplish both.
4. The developer can minimize this problem by stratifying the cost of player actions based on the storytelling arc, based on the rules of the law.
I think this is an incredibly interesting new idea!