Every genre is only one game
Well, sort of. I really mean “systemic game” and I am really talking only about game systems here.
So let me preface this by saying that this article’s title is hyperbolic exaggeration. It uses the term “game” in my annoyingly formal, reductionist way. But I want to say it anyway, for the sake of the provocation; framing it this way jars some preconceived notions about terms out of my head. (At some point, I’ll do a post here about finding alternate, less loaded terms. But for now, since I want to get this out, I’m running with it.)
If you take as given that a game can be analyzed in terms of its grammatical structure — the verbs, nouns, adjectives that make it up – then it leads to the natural thought that you might get the same structure with minor variations.
- This is a rose.
- This is a blue rose.
- This is a red rose with whitish leaves.
- This is a thorny rose with a strong aroma.
A rose by any other name would smell as sweet; a rose is a rose is a rose.
And an FPS is an FPS is an FPS.
(Is this reductionist? Absolutely. It discounts all the things that sit on top of the same skeleton and make them radically different player experiences. But bear with me a moment).
In a game grammar sense, the verbs are the actions the player can take: smell, pluck petals, etc. The nouns are the objects that exist with the system: the petals, the pistils, the stem, the thorns. The rules arise out of the range of possible verbs and the reactions the rose may present to your action. And the “content” – well, that is the statistical variation between roses.
Now consider the small game of RPG combat. The verbs are the actions the player can take: kick, stun, nuke, etc. The nouns are the objects that exist within the system: the sword, the armor, the loot, the orc. The rules arise out of the range of possible verbs and the reactions the orc may present to your action. And the “content” – well, that is the statistical variation between orcs.
But content does not generally affect the rules. It is the x in f(x). The algorithm remains the same, even though the experience may differ radically. We use adjectives – sorry, content – sorry, statistical variation – in order to ring changes on the “machine” that lies underneath.
This means that you can find isomorphism between games. We often identify these as systems held in common, or as “design patterns.”
But often, we can look at a whole constellation of games and say “these have a significant degree of isomorphism.” In some cases, the differences lie entirely in the content. In other cases, they lie in rule variations that are on the face of them minor. We can usually spot these because we call them “a variant.” Poker variants, for example, share 80-90% of their graph, and the often significant differences in dynamics arise because of fairly minor rule changes.
If we choose, we can even look at many of these rule changes as actually being statistical variations. For example, there is a difference in perspective that comes from saying that the number of publicly shared cards in vanilla 5 card poker is a scalar variable that happens to be set to zero. To a player who only knows vanilla poker, the addition of face up cards seems like a new rule, when in fact it’s just like showing that sometimes monsters can group in an RPG, or that gravity can change, when you have thought that in practice gravity was inalterable, or grouping did not exist.
That isn’t really an honest way of looking at it – when a Texas Hold ‘Em is developed, it’s likely that no one approached it as “let’s increment this zeroed out variable to a positive number.” But in analytical retrospect, the addition of a new statistical field could almost be thought of as retroactive to all earlier versions of Poker. And thus poker’s base definition grows to encompass “a game where there might be publicly shared cards.”
As genres develop, they complexify for a number of reasons. Players master the dynamics of the previous system, according to their ability and inclination. Players who have reached mastery demand the same system but with greater variations or better balance, and the response of designers is to add new rules. Special cases arise, like the offsides rule in soccer. Eventually, the ruleset grows rococo, the number of variables extreme, and the accessibility of the genre suffers. A given variant happens to sit at the sweet spot of complexity and accessibility, and becomes the “genre king,” to use Dan Cook’s term. In poker, that was Texas Hold ‘Em.
Then the genre often dies as it spirals off into increasingly rococo variants for an increasingly elite hobbyist audience.
There is a very real craft difference between
- Merely providing statistical changes to a ruleset. This is level design, it’s making new monsters, it’s adding twelve new guns or five classes.
- Providing a whole new statistic that effectively creates a new rule, and thus a host of new mechanics. (Technically, I’d probably only call it a new rule if there’s a new verb to go with it; otherwise, it’s just a new statistic).
- Providing new rules and mechanics altogether, that connect to the existing structure: new verbs and nouns.
- Providing a completely new system of rules: verbs, nouns, everything.
This may seem like pure intellectual wonkery, but it’s not. It’s actually a great aid in thinking about new gameplay, new markets, derivative work, and ambition and innovation, and one of my preferred things to point at when thinking of what game grammar can do in a concrete sense.
From a purely practical point of view, these four tiers serve the following market purposes:
- Providing level packs, more of the same experiences, ongoing support to a fan base for a game genre.
- Adding a minor variant to the game. Likely to retain users into the genre for a longer amount of time, but moving it further down the road to over-complexity.
- Adding a major variant that many will consider a new game altogether, because the variant is extreme enough to create a very different experience: capture the flag versus deathmatch.
- Invention of an actual new game, a new genre. In the long run, may well lead to enormous new markets… just be aware that in practice, this is the hardest thing to do, and also you as the inventor likely won’t be the one who capitalizes the most on it. It’s deeply unlikely to jump right to the genre king, though we have seen it happen (Will Wright seems to have a knack for it).
From a craft point of view, these also fall nicely into system design difficulty tiers. It is harder to do #4 than #1.
Right now, I increasingly see the major variant move, which is exciting. The way in which dimensionality was established as a new platformer variable, for example, has opened up a new fertility in that genre. And this is precisely a formalist sort of move, driving innovation. Identify a variable, see how it can be integrated into an existing superstructure, reap an IGF nomination and some really cool new gameplay.
Ah, but remember I said bear with me regarding the annoyingly reductionist nature of this? You don’t need to approach things from the formalist, mathy end. One of the easiest ways to tackle the hardest task, that of inventing a new genre, is to start with a new experience, one never represented in a game before, and attempt to construct the model for it. The game that makes you feel like a tree, the game about the death of your child, the game about theoretical debates… in other words, to me this points one way towards developing new experiences.
Sometimes, you’ll end up with something that turns out isomorphically identical to an existing game. The FPS becomes the photography game. You may have failed at inventing truly new gameplay, but you have still opened new audiences, reached new expressive potential. (In fact, how come nobody has gone back to photography games and tackled stuff like focus, depth of field, exposure time, and so on? Missed opportunity!)
The biggest thing I think this tool reveals, though, is how incredibly rare it is for new “games” to be invented. As soon as we say that really, there has only been one FPS and variants, one platformer and variants, what a change in perspective!
To me it conjures up vistas of the millions of systems and models we have yet to discover, and the amazing future that awaits us in game design.
33 Responses to “Every genre is only one game”
Sorry, the comment form is closed at this time.
@raphkoster Exceptional essay. I think your point is proven (or more easily seen) in current mobile game creations – match 3, endless runner
RT @raphkoster: Blog: Every genre is only one game: http://t.co/BZKEmPFrWU
Lisa Galarneau liked this on Facebook.
Imagine if the Fatal Frame series incorporated more advanced photography mechanics like DOF and shutter speed. HORRIFYING! But probably also super fun. It’d be like the ARMA of photo games (an admittedly small genre to begin with). haha
Kurt Stangl liked this on Facebook.
Ryan O’Sullivan liked this on Facebook.
I’m not really sure how you define/understand variants or “new statistics”. So, you argue that there there is one FPS and variants and one platformer and variants… But depending on how you define “statistic” and the verbs…one could easily consider one to be a variant of the other – just closer or further from a prototypical game from either genre. It seems too messy… So yes, FPS becomes photography game if you take away the shooting, but FPS becomes 3D platformer if you change shooting for jumping. And then, if you change 3D to 2D – you’re at platformer. Even in the case of “new” games – I’m pretty sure you’ll always be able to trace them easily back to another game… I’d bet that there’s some sort of six degrees of separation for games. How many steps from Doom to Mario? Not that many…
Consider the text-adventure version of Doom (Foom!) here- http://doom.wikia.com/wiki/FOOM!
How different is it in terms of verbs/nouns and adjectives from the original FPS?
There’s also the next 1D “variant” of Wolfenstein3D (playable here: http://wonder-tonic.com/wolf1d/).
@raphkoster I believe the word you are looking for is “ske-lud-ton”. See what I did there?
I clearly am eliding too much of game grammar.
A statistic or content is a numerical variable, in the sense I meant. It is never a verb. Verbs can act upon them. They are parameters, though, not objects.
To go through your examples:
No, FPS does not become photography when you change the verb. The verb has not changed at all. What has changed is a visual representation of the verb — it’s all in the skin.
FPS becomes 3d platformer if you change shooting for jumping — no… that’s a verb change. And FPSes mostly have jumping already anyway. 🙂
The Doom to Mario example definitely involves changing more than the statistical variables. 1d Wolfenstein also removes a host of verbs.
The first two sentences really took the fun out of trolling that, dude.
I understand why you felt compelled to apologize for a potentially controversial blog post title. But that inspired a branch to me.
When I think about the business of game development, the path from invention through labeling into retroactive categorization often does in fact derive from one genre being one game.
Or “Like Spaghetti Westerns, but in Space!”
Game designers know of the nuances between Battlefield from Call of Duty. But many in the business side of this and ancillary industries see “FPS”. For years it was “MMORPG” means “EQ1” until it meant “WoW”. RPGs splintered into “action” RPG and “traditional”. The games that lead to Farmville didn’t have a category until Farmville’s success retroactively provided.
The need for elevator pitches is highly reductive. Good or bad, it’s a requirement in some key parts of this business. But this is a byproduct of a single game achieving the apex of an previously-defined genre, and thankfully not usually the cause.
Maybe not the point you were shooting for 🙂 But in this way, every genre *is* only one game.
@raphkoster Raph Koster in “content is just statistical noise” shocker! Artist & LD jihad imminent.
Heh 🙂
So, in effect, WoW was the same game as UO, but presented a different experience due to…what?
WoW and UO have pretty different verb set overall, and very different rules. Given the complexity of both games, I am pretty sure this lens won’t hold up all that well looking at the entirety of the two games.
That said I think it holds up pretty well if you look at individual games within them, such as the combat systems.
Jason Asbahr liked this on Facebook.
That surprises me. I’ve always felt that they are very similar in your verb sense. In both you gain ability, one called skills and the other levels. In both, you get bonus to those abilities as you progress, it’s just a case of what those bonuses are (expanded greatly in WoW). It’s the rules that separates them and makes for entirely different experiences (this is what I was looking for, I think). The rules are much more strict in WoW, and the result of that is the progressive zones and content. And that is what makes the experience different, at it’s core. While UO’s rules were set up to be more flexible and thus allowed more freedom of choice, and other rules were added on top of that principle.
That’s the way I’ve always seen it, anyways. Wondering what you think of that view.
Of course, you’re talking predominantly about video games in terms of sweeping genres, with variable champion exemplars that do change, but more often create the initial form itself. GTA clone evolves into Sandbox, which then gets loosened/adapted/modified/ignored. However, these leaps into the ‘new’ are quite rare, requiring a previous environment of ‘wash, rince, repeat” to engineer the circumstances allowing the novelty to blossom. The poker example notwithstanding, tabletop games also suffer (?) from cloneitis (Dominion == Deck Building Game, Caylus == Worker Placement game, etc) but are less hidebound, because different mechanisms/techniques can be mixed and matched more easily than their purely digital cousins.
Having said that, I personally had no problem “bearing with [you] regarding the annoyingly reductionist nature of this” until the end. We shouldn’t discount the necessity for incremental changes as necessary for the big leaps. Richard Garfield’s Magic:The Gathering (MTG), possibly the single most successful novel invention in card games (financially and in terms if new game play) needed Strat-o-matic Baseball and Cosmic Encounter to inspire his first, more derivative Five Magics game, which ultimately led to MTG. I’d argue that the really new games, which I agree with you ARE out there waiting to be discovered, are more likely to to be found by those with an eye for the edges of our existing ludographical terrain as well as the willingness to jump into the dark.
I absolutely think this holds true for analog as well. And sports too.
Basically, it’s not really the way the grammar works. A valid way of looking at it, but not the same.
The most basic thing would be to ask “what is the goal a player is told to strive for?” Right off the bat, the two games are very different; WoW puts level as central. UO does not. The range of game loops in WoW is smaller than in UO. The array of objects is far greater in WoW, but I actually suspect the array of types is smaller. I haven’t logged into WoW in close a decade though.
It would be far easier to point at Runescape and UO and say they were the same. And it’s far easier to point at WoW and EQ and say they are the same.
Great observation, Darniaq 🙂
How dare you attempt to define the word ‘game’ precisely. Don’t you understand that you’re silencing the voices of the queer zinesters and indigenous tree people who use it differently? Help! I’m being oppressed! Hate crime!
It’s amusing that, in formal biological terms at least, a rose is most assuredly not a rose, a rose, a rose. That’s a platonic illusion we use for the sake of easy discussion and comprehension, not something that is actually an accurate representation of reality. Taxonomy is a purely abstract system serving our need to categorize and, in your words, ‘Chunk’. Practical biology understands that it has little, if any, bearing on how genetics act in real terms, because genetic diversity acts like a fluid- it is amorphous and defeats separation into chunks.
I think this argument has a similar problem. It’s nice conceptually, but it doesn’t actually help us do much other than pigeonhole objectives into easily described categories. The very fact that the system is used inclines people to thinking in certain ways, which is why the ‘new-game’-is-just-another-FPS phenomenon occurs so frequently. We’re used to thinking in categorical terms, taking sets of parameters as chunked entities and fiddling with one of the parameters for our enjoyment, rather than tinkering with every possible parameter individually. The more we apply such categorization, the less we are capable of conceptualizing outside it. I’m honestly not sure whether this can be avoided, in practical terms anyway, but I think it’s worth noting that this isn’t really a change in perspective, it’s how we already chunk things cognitively and it ultimately reduces the scope of vision, rather than expanding it.
I think it’s far more amazing to observe that, in reality, every individual is their own species. Every living entity is intimately intertwined with every other. Any attempt, however necessary, to place categorical boundaries is ultimately a disservice to the infinite variety and possibilities that genetics throw up. The same is true of game design, perhaps. Categorization is necessary, but on a deeper level, it is not good, because it takes away from the fact that every individual variation of the taxonomic templates we make up is in fact a real thing, with real connections to reality that defy taxonomy.
I don’t want to take away from your model, which is particularly valuable as it does accurately model how people think about games and gives it a language, just as taxonomy maps how people think about creatures and gives that a language. Yet I wonder if you are satisfied with me comparing it to taxonomy, which in terms of actual practice is more or less the bookkeeping service for biologists and geneticists at the coalface. To me, at least, it seems the elegance in your argument is in describing just how unrealistic our conceptions of games really are and how they fail to map the actuality of an individual instance of design.
Yes! Exactly! All these formalisms are, and that is all they are.
Its virtue lies in the fact that it is a different lens than the one in common currency. Yes, it causes us to look at things one way. But when we have grown used to looking at things another way to the extent that we can’t even understand this post (which I have gotten a LOT), that has value.
An example: I have just seen someone pointing to Tim Rogers’ article on friction (http://kotaku.com/5558166/in-praise-of-sticky-friction) as a great design article (which it is) and as “the most important article written about how to criticize games,” which it isn’t. It focuses entirely on game feel. Game feel is incredibly important. But it helps not at all with inventing new genres.
And deeper down, seeing everything as unique enforces barriers between them and pushes us away from seeing commonalities that are important, and so on, turtles all the way down. It’s important that we be able to shift fluidly between perspectives.
>>Its virtue lies in the fact that it is a different lens than the one in common currency. Yes, it causes us to look at things one way. But when we have grown used to looking at things another way to the extent that we can’t even understand this post (which I have gotten a LOT), that has value.
But is it really? I mean, in terms of discussion, sure, but in terms of the way people actually look at and categorise games? I get the very distinct impression from my talks with gamers and game designers that they kind of innately do what you’re suggesting here- they have a paragon game that to them represents what X genre means and they relate other games to it. Whether a game can be called a new genre is not how it differs from the aggregate of all games considered part of that genre, but how much it differs from their own personal paragon game. It feels to me like this is more of a way of coming out and formalizing a thought process that many (perhaps most) people use already. Of course, our experiences could be vastly different. I don’t get to talk to industry folks as much as you, that’s for sure.
>>It focuses entirely on game feel. Game feel is incredibly important. But it helps not at all with inventing new genres.
Why not? There are plenty of real-world games that revolve around game feel- egg and spoon races, stick your hand in the box and guess games, etc. to divorce the abstract objectives and restrictions of these games from their ‘feel’ would entirely miss the point of the exercise. I appreciate that in videogame design there are differences, but the very existence of those non-digital games suggests to me there’s plenty of opportunity for mechanical development specifically around kinaesthetics. In fact, I would argue that focusing on game feel is precisely an area that is ripe for producing new genres. I mean, I totally agree that the article isn’t the most important ever, but to say it has no potential for creating novel verbs, as you called them, is… a little hasty. We’ve got plenty of verbs dealing with visual, audio and mechanical interaction, but precious few dealing with kinetic interactions.
>> And deeper down, seeing everything as unique enforces barriers between them
I very much disagree with this. All things share connections and commonalities at some level, even if it’s at the atomic. Understanding something’s uniqueness must incorporate an understanding of all the things that make it the same as other things, as how else can you find out what in particular distinguishes it. Uniqueness is only ever a property given by a large combination of properties which are shared with many other things. Understanding a huge field of those common properties and looking deeper and further to find more and more is really the only sure way of being able to build something unique yourself.
RT @raphkoster: Blog: Every genre is only one game: http://t.co/BZKEmPFrWU
RT @raphkoster: Blog: Every genre is only one game: http://t.co/BZKEmPFrWU
Formalism is basically the practice of taking a set of data points and making a linear regression out of them. If you’re lucky, that line passes through a number of the points. But by being able to describe that line, you’re able to make much more precise descriptions about each of the data points themselves… or even if the precision doesn’t increase, it’s at bare minimum a way to relate two data points to each other usefully. This might not be required for something as difficult to quantify and infinite-dimensional as experiences or even genres, but it’s certainly not devoid of value.
Formalizing a thought process that we use already is standard practice for going from casual usage to academic rigor. Everything does that: logic, music, painting, photography, etc.
I’m very curious: would you say that the “german-style board game genre” is actually composed of many different games, unlike the “FPS genre”?
Yes, I would. I am not a huge German board game fan, but every one I have seen is a different game in this sense.
Hi Raph,
I wrote this article a while ago, and I wonder if you like the way it tries to capture verbs and nouns and the relationships between them. I use a portion of Hjelmslev’s language theory in my piece. I think that it deals with the same problems, but it looks at the issue from a different perspective.
http://gamasutra.com/blogs/AltugIsigan/20121020/179842/Rock_Paper_Scissors_A_Linguistic_Approach.php
We never really deal with the whole because we are part of that whole. To grasp the whole in its entirety would be an infinite regress — we’d have to contain the whole and with the whole ourselves who contain that whole and and so on ad absurdum. So we have no other choice than to cut the whole into pieces according to our tastes (i.e. value judgment is constantly at play everywhere.) The very fact that we see ourselves separate from the rest of the world is a value judgment — we aren’t really separate from anything or anyone. When we cut the whole into pieces we create a fiction, a map, a theory, that describes what repeats and what doesn’t repeat. We then use that map to explore the territory to its limits (in our case, the territory is the artform.) The map then undergoes many revisions as we discover new things. And, well, that’s it. The point here is to understand that it’s impossible to explore anything without a map and that all mapping is in effect value judgment.
By the way! terms such as “formalism” and “reductionism” are ugly and superfluous. I’d simply use “theory”.
[…] •Mathematics – symmetrical sets of numbers, their graphs, and transformative relationships. (You’ll find much of the latter in the artwork of M.C. Escher) •Biology – similarity in body structures or evolutionary paths between species. •Linguistics – common evolution of word origins, usage, and grammar. (See: Zipf’s Law) •Game Theory – decision trees which produce mutual stability points, such as Mutually Assured Destruction, a form of Nash Equilibrium – the best-known work of John Forbes Nash •Chemistry – molecules that share structures and bonds. •Sociology – repetition of group behavioral and social structures. •Psychology – relations between stimuli, actions and memory; Between subparts of the mind and the whole. (See: Hofstadter’s Pullitzer-Prize-Winning Gödel, Escher and Bach) •Software – programs that use common sections of code or mechanics (i.e. different versions of Linux). Raph Koster explains game engines this way. (See: Every Genre Is Only One Game) […]