Gamasutra: The Art & Business of Making Gamesspacer
arrowPress Releases
August 21, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

The four lenses of game making
The four lenses of game making
January 6, 2012 | By Tadhg Kelly

January 6, 2012 | By Tadhg Kelly
More: Console/PC, Social/Online, Smartphone/Tablet, Indie, Design

[In this opinion piece originally posted on the What Games Are blog, and reprinted in full with his permission, UK-based game designer Tadhg Kelly breaks down the four lenses of game making, a "common set of assumptions and predispositions" he often sees in developers.]

For years it's been apparent that interpreting games and their makers through the opposed lenses of gameplay or story is inadequate. Such a one-dimensional spectrum breeds false oppositions (fun-or-art?) while either ignoring many games that don't fit or reinterpreting them so they fit badly. The spectrum is too reductive and, while it is easy to summarize, it leaves out too much context.

Rather than talking about games in terms of two lenses, I use four (potentially five, but I'll come back to that). Each represents a common set of assumptions and predispositions that I often see in makers, and there are correlations between them which makes for an interesting (though perhaps deceptively symmetric) diagram.

This post is long, but I'd like to take you through each in turn. I think you'll find it useful.

Lenses and axes

A lens is a cognitive, emotional or perceptual bias which affects how information is received, understood and contextualized. In politics, religion, philosophy and the arts there are many lenses, and game makers have their lenses too. They have different ideas on the role of the player in the game, or the role of rewards. Some believe that structured goals matter, others that player self expression is king.

The lens through which a game maker sees games tends to affect the kind of work that they consider worthwhile or worthless, and I suspect is why most makers tend to concentrate on one type of game over their career. Lenses therefore describe categories of games (or modes within games) as well as their makers.

In the dual-lens model of gameplay-or-story the relationship between the two is often described as an axis. Similarly, I describe lenses by two axes, leading to a 2x2 grid. Each axis represents an interest, and so each lens represents a combination of two interests. This is similar to Richard Bartle's description of player types, and the resulting graph shows four clear lenses, with room inside each for wide variations.

They are:
  • the Frame Axis (Emergence-Experience)
  • the Fantasy Axis (Role-Rule)
The frame axis

The Frame Axis is about the importance of uncertain or certain outcomes, and therefore whether the game is designed toward emergent or experient play.

The frame of a game is its mechanical layer. It is the levers and environment of the game stripped of all context and aesthetic concerns. Everything in the frame of the game boils down to binary information (allies/enemies, dangerous/helpful, win/lose etc) and the frame is the only part of a game that the play brain understands.

A frame is more inclined toward emergence if it permits a high degree of discovery, especially of the kind that the game's designers never foresaw. Unusual strategies, innovation and creativity are enabled, and in some ways the game maker feels as much a participant in the discovery of the game as the creator of it. The objective of developing an emergent frame is a system in which a limited set of actions and rules produces a near infinite set of outcomes.

A frame is more inclined toward experience if it delivers predictable emotional engagement. The game is planned, measured and assessed for impact. It pulls players along in well understood patterns, and is unconcerned with player innovation. The objective of developing an experient frame is a system that manipulates players and leads to set piece outcomes that they find compelling.

Robust rules are most likely to produce emergent play, and so makers who like emergence tend to design frames with elegant engines that encompass all possibilities. Experient design is less concerned with overall robustness and more on moments. They often have frail frames, meaning that they have limited ranges of available interaction, but smooth over this by pushing the player toward key decisions, playing on their sense of anticipation and using theatrics to draw their attention.

Most games are neither wholly emergent or experient. The majority use a bit of both, favoring one overall.

The fantasy axis

Games draw players into other worlds and empower them to take decisive action within their confines. However worlds vary wildly in how they are presented. The Fantasy Axis is about abstraction versus fidelity, nakedness or richness, and whether the game gives the player a strong contextual role, or regards role as secondary to the formal rules of play.

If the frame of the game is its mechanical layer then the fantasy is the creative layer of art, sound, text, numina, animation, fiction and so forth that sits on top of the frame. Fantasy gives the game world an identity, from the simple and iconographic through to lush realism. In encourages empathy, and communicates to the player culturally as well as intellectually. It might be realistic or stylized, hand-crafted or generated procedurally, narrative or open-ended.

Fantasies which tend toward rules are formal, literal and unambiguous. The frame is highly visible and the player simply plays as herself, the invisible hand that causes change. Strongly rules-oriented games might only use elements like numbers, cards, squares and shapes and have only a few distinctive terms that define conditions in the game.

Fantasies which tend toward roles want to hide the frame in favor of painting a landscape. While the player is still taking action to win, she does so within a convincing and elaborate context that tries to define an identity for her. Her role is a job description, a part of a cast within a world, even possibly a persona.

Strongly fantastical games therefore tend to spend a lot of time on characterization, visual elements, music, voice-overs, special effects and language. They de-emphasize the frame as much as possible (sometimes too much) in order to induce the art brain to believe.

Tetrism (emergence, rule)

Tetrism is about creating neat game dynamics that are consistently fun and inviting the player to master them over a long term. It is inspired by the elegance of many classic games, like chess, bridge, crosswords or soccer.

Tetrist games are formal, based around a few defining and highly extensible actions and bounded by rules which bring the game toward an inevitable conclusion (such as time, points, speed or increasing difficulty). Although moderately tetrist games are often charming (such as many indie or casual games), the frame is both clear and emergent, and the win conditions are usually victories.

Tetrist games are tests of skill and strategy. Their game dynamics are often easily described, such as sorting blocks, making words, kicking a ball or moving pieces on a board. Story is perfunctory or non-existent. Optimal tactics and strategy are high.

The value of tetrism is its focus on the kind of engagement that the play brain thrives on. Tetrist games are usually the ones that break out on new platforms to stun the world with simple genius and tetrists are always keen to see what new devices bring opportunities to create novel dynamics. However they are not just motivated by new interfaces: Even on older platforms like the PC there are many new tetrist games every year.

Tetrists often think of themselves as the most pure kind of game maker. They aspire to find that unique innovation or invention that will spawn a generation of games, and so they are often keenly aware of (and overshadowed by) past games and their successes. This leaves them feeling left behind by a game development culture that spends a great deal of time on complicating what tetrists feel should be elegant and universal. They also tend to react negatively to ideas such as free-to-play economics because of the perceived issues of fairness.

Possible examples of tetrists might include Shigeru Miyamoto (Mario), Alexey Pajitnov (of Tetris fame), Jeff Minter, Kyle Gabler and Ron Carmel (World of Goo), Andreas Illiger (of Tiny Wings) and Adam Atomic (Canabalt).

Narrativism (experience, role)

Narrativism is about using a game to impart a storied experience in which the player takes an active role and develops sympathy toward its outcomes. It is inspired by literature, cinema, theatre and other narrative arts, and places videogames as an inheritor of those forms.

The narrativist wants the player to feel more than just the joy of winning. She wants him to care, on a personal level, about what happens to the characters in the game, and to experience sensations of loss, hope, and sadness as well as thrills. She wants the player to feel as a dramatic hero in his own play. Narrativist games aim to be aesthetically coherent. Their makers spend a lot of time creating the place, people, sights and sounds of their world. Richness, authenticity and production values matter. Back story, characterization and theme likewise.

Moderately narrativist games combine mildly emergent gameplay and opportunities for discovery with outcomes that are fairy predictable. The result is storysense. Strongly narrativist games move away from storysense into storytelling, attempting to characterise the player and limiting emergence in favour of experience. They often become opaque or are easily mastered, leading to unintended boredom. At their most extreme, narrativist projects abandon the idea of 'game' altogether and becomes a non-game, like a virtual promenade or amusement ride.

Possibly the biggest issue for narrativists is validation. They often consider themselves to be artists and aspire for their games to be taken as seriously as cinema on the global cultural stage. Some even believe that games will one day eclipse or eat all other forms of art and contrast interactive art as somehow 'better' than so-called passive art. Like all institutionalists, they are essentially looking to an art world to confer legitimacy upon them.

Possible examples of narrativists might include David Cage (Heavy Rain), David Jaffe (God of War), Tim Schaeffer (Grim Fandango, Psychonauts), Auriea Harvey and Michael Samyn (Tale of Tales), Jesse Schell (The Art of Game Design) and Jason Rohrer (Passage).

Simulationism (emergence, role)

Simulationism is about creating an endless and authentic world in which a player can explore, experiment and discover. It is inspired by artificial intelligence, holodecks and the infinite possibilities of software itself.

Simulationist games have complex rules and robust systems underpinning them, but their makers do not like those systems to be visible or bounded. Wins, tasks and formal goals sit uncomfortably within a game that tries to create (or recreate) a plausible world, as does direction of emotion. A simulationist wants the player to be as self-directed as possible, to feel the delight of surprise or meaning on their own terms.

Simulations cast the player in a strong, but easily described, role and then build the entire world around making that role feel real. Roles can be as varied as pilot, adventurer, ruler of city or state, rail driver, wizard, god, theme park owner, football manager, evil genius and so on. They focus on making complex effects tangible to the player and conveying a sense of consequences for all actions. Simulationist games also tend to encourage creative play, with players using the game itself as a canvas.

In moderate simulation there is usually some amount of formal structure and direction to the game, especially through its early stages. The best games of this type include a blend of true simulation with simulacra (faked simulation), with the overall effect that the game is less than real but more easily grokked.

However in strong simulations everything is emergent, procedurally generated and aims for infinite possibility. Simulationism's biggest pitfall is therefore opacity. Worlds can become too complex, roles too indistinct and complex effects too subtle for players to perceive, and far from appreciating the genius of the engine powering this world the player often has adverse reactions: the game becomes random, unfair or frustrating.

Simulationists often consider themselves to be at the cutting edge, like research scientists. They believe that simulation is the point of videogames, and that the art and individuated play of games are one and the same. Therefore the truth of the game is what matters most, and simulationists are usually disappointed by games that they regard as faking, manipulating or spoon-feeding players.

Possible examples of simulationists might include Will Wright (Sim City, The Sims, Spore), Peter Molyneux (Populous, Black and White), David Braben (Elite, Kinectimals), Sid Meier (Civilization), Todd Howard (The Elder Scrolls), Markus Persson (Minecraft), Chris Delay (Darwinia), Andrew Stern and Michael Mateas (Facade) and Kristoffer Toubourg (EVE Online).

Behaviorism (experience, rule)

Behaviorism is about considering the player as a collection of desires and creating systems that satisfy those desires. It is inspired by behavioral and motivational psychology, and considers all games as challenge, anticipation and reward engines.

Behaviorists model their games on psychological hooks that open loops, draw engagement and encourage emotional attachment to outcomes. They use repetitive actions to complete those loops and deliver rewards. The anticipation of a loop's end, and the reward, has a powerful effect on the human mind and can engender feelings of optimism. In older forms that would mean money, such as a slot machine or a lottery. In newer forms it sometimes means points or virtual goods that the player will find useful toward another goal.

The big revolution of behaviorism is metrics. Behaviorists measure everything that they possibly can about their players, test small changes and then measure their outcome. However this means that behaviorists tend to be wary of emergence. Emergence is generally hard to directly measure, and to the behaviorist anything that cannot be measured cannot be reliably improved.

So behaviorists tend to be the most creatively conservative of all lenses. They consider it better to copy a successful game and improve upon it, or adapt another game (perhaps with a different theme) rather than create from scratch. This sort of lean and predictable approach is why behaviorists are the darlings of the investment scene, but also why their games tend to be lower value on a per-user basis than all other kinds. They are far less likely to engender true loyalty or a fan culture.

Behaviorism struggles with the ethics of game making in a way that other lenses do not. Gambling, addiction and exploitation are serious social issues and behaviorist games skirt closer to them than any other kind. On the positive side, many behaviorist game makers believe that games can be agents of social change and self improvement.

They think of themselves as the next generation, characterizing other lenses as subjective or backward. Yet strong behaviorist game design (particularly gamification) tends to flounder when faced with the play brain's ability to figure out dominant tactics and only care about extrinsic motivators (i.e. prizes) while ignoring wider ideals. This more than anything is why makers from other lenses (particularly tetrists) tend to regard behaviorism as either deluded or evil.

Possible examples of behaviorists might include Jane McGonigal (Reality is Broken, SuperBetter), Brian Reynolds (FrontierVille), Dave Maestri (Mob Wars), Mark Pincus (Zynga), Brenda Braithwaite (Ravenwood Fair), Amy Jo Kim (Consultant), Gabe Zichermann (Gamification evangelist), Michael Acton-Smith (Moshi Monsters) and Sebastian Deterding (researcher and designer).

A fifth lens?

Regular readers of What Games Are know that I often talk about the boundaries of games, like:
  • Storysense, not storytelling.
  • Ground your gameplay but don't idealize founderworks.
  • Avoid meta-games.
  • And the Techthulhu dream.
In a sense this blog advocates a centrist position that says that there are lessons to learn from all lenses, but also that the most memorable video games often come from the space where lenses cross rather than the far corners.

Super Mario 64 tends toward tetrism but has influences of simulationism, narrativism and behaviorism. The Sims is a simulation, but aspects of it are tetrist and behaviorist, and there is a sense of narrative. Final Fantasy VII is narrativist, but with simulation in the form of team building and exploration, emergent tactics in combat and some optional economic grinding too.

To their fans, each is a legendary game, clearly thaumatic. Using some of each lens can be synergistic, but I think there's more than simple cross-over effects at work. Elements which pull on each other can just as easily lead to dysnergy (less than the sum of its parts), which nobody wants. There's an x-factor of vision, invention and creative voice in each of these games, and that's not something that sits easily on any axis.

Since this blog invented the word thauma to describe the uniquely magic of games, I might as well call the lens that believes in the importance of x-factors thaumatism.

To the thaumatist, games are an art as they are, not as they might be. There are boundaries to work with (mostly set by players, not designers) and principles to follow. Somewhat experient, somewhat emergent, a bit governed by rules and influenced by role, thaumatism sits above other lenses, blatantly borrows from all of them and unifies them with intent. And that somehow translates into magic.

Is it a lens as such, or just a way of describing how game makers can draw from the founderworks of every lens to create masterworks? That's for you to decide.

[An Irish lead designer and producer living in London, Tadhg Kelly is the author of a challenging book about, as he describes it, "Reclaiming games as an art, craft and industry on its own terms", entitled What Games Are. The blog for the book is You can also follow his tweets on Twitter (@tiedtiger).]

Related Jobs

Gameloft — Seattle, Washington, United States

Creative Director
Yoh — Vancouver, British Columbia, Canada

Rendering Engineer Job
Yoh — Vancouver, British Columbia, Canada

Multiplayer Designer Job
Firaxis Games
Firaxis Games — Sparks, Baltimore, Maryland, United States

Senior Visual Effects Artist


E McNeill
profile image
There's a lot of insight here, but I think this classification scheme (like most) can obscure a lot of important differences. Contrast the aesthetics of "tetrist" games Tiny Wings versus Go, for example. Or the way that "narrativist" games Passage and God of War integrate their stories and mechanics. In some ways they couldn't be further apart.

There are a lot of axes along which we could plot games, not all of them orthogonal. I think it's misguided to present two of these axes together unless you can argue that this presents the most important types of classification. You argue that a one-axis system is "too reductive and... leaves out too much context." I argue that your system has the same problem.

Steven An
profile image
Totally agreed. I kinda chuckled when I saw "A Fifth Lens?"....yeah man, and there's probably a 6th, 7th, etc. etc. etc. :P

E McNeill
profile image
"The frame of a game is its mechanical layer. It is the levers and environment of the game stripped of all context and aesthetic concerns."

The choice of the term "frame" shows a bit of bias, I think. "Frame" implies something peripheral or supplemental, while I (and half of your diagram) would argue that the mechanical layer is the core of the game. It's the aesthetics that "frame" the mechanics, not the other way around.

Tadhg Kelly
profile image
Interestingly, a commenter on my own blog thought I was being biased in the other direction :)

E McNeill
profile image
I'm not trying to argue that you're biased (and I hope my tone doesn't come off as hostile), but rather that this one instance of word choice is biased. Overall, you came across as an extreme moderate. If that's what you were going for, maybe this choice of word is something to change.

Tadhg Kelly
profile image
Ok, thanks.

I'm using frame in the sense of construction, as in the girders, wiring and such. So it encompasses both the physical arrangement of how the environment works (and in a sense therefore it is also what the fantasy is painted onto) and the rules that operate within it.

E McNeill
profile image
I see. That makes sense. I was interpreting it more like "picture frame" or "framing".

Anna Tito
profile image
Definitely lots of food for thought. You do come across as a little pro-emergence anti-experience, but I think that is largely just because the amount of time spent on discussing emergence vs the amount spent on experience. I definitely think the 5th category is an interesting one, part of me wants it to be a category all of it's own another part of me thinks that the category is not really useful as a well balanced, well designed game is a balanced well designed game, it doesn't need a category to define it.

Bart Stewart
profile image
In describing a theory of things, usually developed after a lot of very focused research, it's always tempting to veer off into esoterica and lapse into jargon as a sort of preemptive strike against the inevitable criticisms. Thus it's always really satisfying to read a design theory article like this one that communicates its concepts in simple, clear language that anyone can appreciate. Nicely done!

I (probably unsurprisingly) can't resist noting that the four lenses described here seem to me to be another useful way of expressing the single fundamental theory of playstyles that other game designers have proposed, which I described a few months ago here at Gamasutra as a "unified model of playstyles":
tyles_a_.php . The Simulationist and Narrativist lenses are virtually identical to the Rational/Explorer/Simulationist and Idealist/Socializer/Narrativist playstyles, respectively. And I think a strong case can also be made for the Tetrist lens, with its emphasis on following simple rules for a clear win state, being analogous to the Guardian/Achiever/Gamist playstyle, as well as the Behaviorist lens (focusing on the manipulation of people and explicit references to "gambling" and "addiction") being an analog to the Artisan/Killer(Manipulator)/"Experientialist" playstyle. I think it's also telling that these four lenses, if you rotate Tadhg's chart counter-clockwise 90 degrees, exactly match the playstyles in the four quadrants of the Unified Model diagrams.

I'd be interested in hearing any thoughts, pro or con, about this possibility.

Either way, I agree fully with the final point: a good designer will always look for appropriate ways to tap into elements of any/all of these four styles. This doesn't mean counting up playstyle-specific elements to try to make them equal in strength -- it's OK to have a game whose features emphasize a particular style. In fact, that focus might even be a good idea for most games.

What this approach says is that it's also OK to think about ways in which other people can enjoy your game without diluting it for your primary audience, and to design systems and features that enable these additional ways of enjoying the game. Minecraft (to take one good example) is mostly about exploration in a large procedurally-generated world and creatively building within that world (the Simulationist lens). But it immediately becomes more interesting when exploring that world is risky and multiplayer, attracting Behaviorist-motivated gamers. (And note that just before Minecraft v1.0 was released, elements of Achievement and Narrative were added as well.) You don't want to randomly lard a game with features that appeal to different kinds of gamers, but a judicious selection of playstyle-focused features can, I believe, make the difference between a niche game and a breakout hit.

The thing is, to have that understanding of what gameplay features appeal to different kinds of gamers and how best to design such features into your game, you really do need to think in the kind of "thaumatic," all-four-styles way that Tadhg describes here. It's fine to have a personal preference for building a particular kind of game, but having a practical understanding of all four styles is equivalent to a novelist being fluent in a language. Isn't understanding what gamers like -- and why -- an important part of being fluent in the language of game design?

This article is an excellent addition to the "know what your audience likes and why" school of game design.

Rich Boss
profile image
Simulationist = Artisan/Killer(Manipulator)/"Experientialist" playstyle

Narrativist = Idealist/Socializer/Narrativist

Tetrist = Rational/Explorer/Simulationist

Behaviorist = Guardian/Achiever/Gamist

They key to understanding the link between the definitions is to think about what each type desires. What a person wants determines decision and action. The real question is: How do we determine what a person wants?

Bart Stewart
profile image
Rich, I'm always open to other approaches to this stuff. Why do you think that Tadhg's Simulationist lens is not another form of the Rational/Explorer/Simulationist playstyle, or that the rules-oriented Tetrist lens is closer to behaviorism than to the Gamist playstyle preference?

As for figuring out what a person wants, "ask them" and "watch what they actually do given a real range of options" are the two ways I know of. Are there others that could work?

Axel Cholewa
profile image
As far as I understood the article it's more about games and game designers and not so much about play styles. While there certainly are correlations, mapping one onto the other is ambiguous, at best.

Rich Boss
profile image
Tadgh, you have a wonderful framework for understanding games here. What brought you to the realization that games could be categorized like this?

Louis Png
profile image
Well, I do believe this article opens up another view of game design: game designers themselves.

Great read, helps me understand myself a bit, and by using Bartle's classification, I realized that the kind of designer I am, is quite different from the player I am. (Personality split?)

But I think this article will be a great starting point for desginers, as Sun Tsu said, knowing yourself as well as your enemies is the key to many victories.

Joe McGinn
profile image
Interesing article. Two points for your consideration:

1) Tetris seems almost entirely experiential by your definition. Maybe you could explain it better. Is this not Tetris to a T? Maybe I'm not understanding something.

"A frame is more inclined toward experience if it delivers predictable emotional engagement. The game is planned, measured and assessed for impact. It pulls players along in well understood patterns, and is unconcerned with player innovation. "

2) To be honest the fifth category, thaumatism, weakens the piece. Unlike the other categories there's no strong theory why it's superior, and it's easy to think of counter-examples of great games that don't fit there (Skyrim is pretty firmly in your Simulationist corner).

Good read overall though, food for thought, thanks.

Josh Foreman
profile image
Brilliant work as always, Tadhg.