A bit on how I think games work
Abalieno at The Cesspit reacted kinda negatvely to “The Healing Game” and raised some interesting points. But I think we’re talking past each other to a degree, so I wanted to take a step back, and make sure we agree on terms. The below is the framework that I am using in thinking about “How Games Work,” which I am thinking about a lot because that is, broadly speaking, the next book.
First, there is a part of the game that is the bare mechanics and rules. This part is irreducible; if you take it out, what you have left is no longer considered a game. These bare mechanics and rules form a system that players can interact with; poke and prod the system, and you will get back various responses. The player is not in the system, but they might have a very evolved proxy that is (such as a character in an RPG).
Then there is what might be called “content.” It’s statistical variations on the system. It’s slightly different input provided to the same rules. Keep in mind here we are still oeprating fully in the abstract.
Then there is the presentation of the game. This is not at all a simple subject, either, and it is tremendously important. Games without presentation barely exist (there are a few, such as playing chess entirely in your head). Now the mechanics are called “combat” and the statistical variation is called “monsters” and so on. Basically, what we have is a metaphor that encompasses the mechanics and the content.
Above and beyond the presentation is another layer of what typically gets called “content,” which includes things like narrative. Now we’re into familiar territory, so I won’t elaborate too much on this, save to say that instead of being statistical variation on the mechanics, it’s variation on the metaphor.
Abalieno is arguing that the healing example is bad because game design should always proceed from the narrative end. My example proceeded from the mechanical end.
I am going to insist that both approaches are valid. Many many games have been created that are classics that started from the mechanical end — particularly in the realm of boardgames, but also in the arcade with things like say, Tempest or Qix. And many are the classic games that started from the narrative end. But perhaps most telling are all the games that originated somewhere in the middle — one can easily imagine Joust starting there, for example.
- Were I to do a game “about healing” I would be starting with the metaphor.
- Were I to do a game about wiping disease from the face of Africa, I would be starting with the narrative.
The best games will achieve a marriage of mechanics and metaphor to the degree where you won’t be able to tell where it all began; both will be considered holistically. (They will still be susceptible to the above analysis, however).
Seen in this light, we have much to examine and think about in each of these layers.
- At the mechanical layer, we have issues such as how to design game systems, the thoughts on game design grammars, much of the thoughts on ludology though not all, and perhaps most importantly, the discussion of innovative gameplay.
- At the statistical variation level, we have questions such as what is “balance,” emergent behaviors, repetitiveness, game scope, levels, and so on.
- At the metaphor level we can have thought experiments such as “The Healing Game” and discussions of whether or not our metaphors are the right ones.
- Finally, there’s a ton for all of us to learn at the narrative layer, from maximizing our learning from other media to mastering the tools of storytelling, of art design, of architecture, and so on.
This breakdown is, perhaps, why game designers must be multidisciplinary. After all, the layers correspond roughly to science, mathematics, philosophy, and creative arts. To make something that works holistically, you’ll need to draw on all those ways of looking at the world.
The one that feels irreducible to me, of course, is the core mechanic. That is what makes this medium be games, and not something else.
90 Responses to “A bit on how I think games work”
Sorry, the comment form is closed at this time.
Tell me more about this cheddar-powered radioactive behemoth. HELLO ME NOT DEAD Heh! Pastoral marriage counseling: What if? What if Christian married people were actually expected to know, remember, and honor their wedding vows?Interesting game idea Using Texas Hold-Em as a fight mechanic in an online game. Women?s Health and Human Life Protection Act Good news for unborn babies! You are now considered valuable by the government of South Dakota.
I agree with the post, but it should be pointed out that one of the most important purposes of any given layer is to provide “angles of approach” on the other layers. Your mechanic is explored by the variations, which are created by the theme, which is created by the narrative. Similarly, your narrative and variations are limited by your gameplay, your theme is limited by your variations and narrative…
All the layers establish the “width” of the other layers. Using the exact same gameplay, you get dramatically different variations by changing your theme.
This is why people complained about your “healing game” that kept the game dynamics the same: changing the theme means that there’s a difference in the variations which are available. Many people recognized this instantly. Changing the narrative – such as a game about curing Africa of AIDS – radically changes both the theme and the variations… even if the gameplay is the same.
I hope I’m being clear. All I’m really trying to say is: “All these layers are separate, but each guides the others.”
A profound post. And I’m not one to normally shower praise! I found equating the layers to the different disciplines to be particularly eye-opening.
One thought to add: Craig’s comment speaks of “angles of approach” from one layer to another. That might be one way to think of it. Another way to think of it is that the different layers may act as filters – limiting the perspective on, or choices in, another layer. In some cases that might be an actual limitation (e.g. game mechanic or ‘balance’ decision limits what a user can do) or a perceived limitation (because of the narrative’s basis in reality, preconceived notions exist about what’s possible and what’s not – “Why didn’t you just drive the car up the side of the building?” – “Cars can’t drive up buildings!”)
Maybe Craig and I are saying the same thing, but I’m not so sure. Filtering is one layer limiting choices in, or views of, another. “Angles of approach” maybe be more about the presentation of a layer to the player in a certain way.
Kim: I think we’re talking about the same thing. A filter doesn’t just block: it also highlights. Nice clarification, thanks.
Hmm.. Hmm..
I’ll have to think more later about this. I read it as if those four steps work in a hierarchy, where you are free to choose one and put it at the top, dominating the others.
I’m not convinced by the last two because you bring the example of your “healing game” for the metaphoric level which instead I consider JUST mechanical. And I also think the narrative level blends with it. I find hard to separate the two.
The point is that I see those levels much more dependent one from the other. Your “healing game” instead was a metaphor completely independent from the mechanical layer. So, that example started from the metaphor or from the mechanic?
I see the whole discussion like this: if it started as a pure mechanic it would hardly translate in that type of metaphor. I mean, If I have that mechanic I’m probably going to present it in a different flavor when I go to choose which metaphor is more appropriate. If instead it started from the metaphor, well, again I would design the mechanic to be MUCH different and more appropriate to the metaphor. To achieve better the communication of that metaphor.
Which also means: despite I recognize the hierarchy, sometimes the result is the same whether you start from the end or the beginning (as things are connected).
So, from my point of view, you can take it either way and that example is still something that doesn’t make sense.
If this is true, it means that the mechanical level would STRONGLY depend on the metaphor. Again following the model as a hierarchy where you decide to start from the metaphor and, from there, figure out the other parts. Tabula rasa. You start from a blank paper, set a point wherever you want, and then start to draw the first lines. The rest will be progressively derived accordingly to what is on top of the hierarchy. So everything is connected.
If you see them as hierarchies, I agree that those level exist and they are always ALL present in every game. I also agree that you can freely reorder the hierarchy as you want. We have concrete examples for every possibility.
That said, I believe there’s a “bias” and we are definitely, unavoidably going in a precise direction that is the one of the metaphor (the one I’d say should be higher on the hierarchy). Even the cinema started with the technical experiments to then move to bring up the emotions. I see a definite progression where the mechanical level will be progressively overshadowed and enslaved by the other levels. For example there are “narrative techniques” to obtain certain effects, but these techniques are bent to be functional to the narrative.
Games in general have always been more tightly connected to the mechanical level, also because they started from the *interaction*. But I think that, whether we like it or not, we are going toward the emotional, symbolic level where the mechanics will become progressively hidden. As with the use of computer graphic in a movie, the best case is the one where you don’t see it, where this technical level becomes completely functional to the narrative needs. As I wrote in the post you linked, we are made of symbols.
Let me rephrase: games in general has always been more tightly connected to the mechanical levels, because there’s a definite predominance of programmers and because “game design” has always been considered superfluous.
I believe that many if not the majority of the woes in this industry are imputable to people sitting on the wrong chairs.
The more we’ll see genuine game designers and not programmers-recycled-designers (through the social treadmill called “promotion”), the more the games will start to shapeshift into something else. And the technical constraints will loosen up.
When this will happen games will finally truly become a “medium”. Hopefully shying away from becoming completely autoreferential as 99% of the garbage that arrives on TV and instead telling us something valuable about ourselves and the world outside.
I continue to disagree on this point. Game designers “may” be multidisciplinary. It’s surely useful and helpful. But not “obligatory”. That’s just blindness from my point of view.
Knowledge can help but it isn’t everything, nor what is truly important. As you don’t need to go to a writing school if you want to become a writer. I just refuse to codify this, there are many different approaches and the game designer with the most knowledge isn’t going to be the one univocally making the best games.
He’ll have an advantage, but I wouldn’t give that advantage a fundamental role.
abalieno wrote: “As you donāt need to go to a writing school if you want to become a writer.”
No, but you do need to know how to write. I don’t think Raph meant multi-disciplinary mean “a degree in each”. Just “well versed”.
Not to put words in Raph’s mouth, but perhaps he meant they’d have an advantage if they are multi-disciplinary? But then I’d argue that’s true for just about everything in life š
Yes, I meant that it’s an advantage to be multidisciplinary. Sorry, didn’t mean to sound quite so absolutist!
Abalieno, when you say hierarchy, to me that means an order. I don’t see these as having an order, save perhaps for the one irreducible thing, which is the mechanics. No mechanics, no game. This is where I usually use my analogy to dance; no dance moves, no dance. You can then have lots of other stuff (lighting, costuming, etc, all the stuff I listed in the book) but those things can all exist without there being a dance.
But everything else — I had not thought of it in terms of a hierarchy. I suppose that systemic statistical variation cannot exist without there eing a system, and I suppose that a narrative cannot exist without there being a metaphor. So at that point we’re back to the good ol’ ludological vs. narrativist split.
Broadly speaking, I suppose you’d say I come down on the ludological side, because I do grant a certain sort of primacy to mechanics. That’s why I tend to call everything else (which isn’t just narrative stuff) “the dressing” — it’s the stuff that orbits the nucleus, which is the game mechanics.
But that doesn’t mean it’s unimportant. Because the thing built is no longer just a game (which can exist with just mechanics and no metaphor, as Go does, for example) — it’s an experience. That’s why you can create a radically different experience by using the same nucleus.
Now, the point of the healing game exercise is to demonstrate that a given nucleus can use many different metaphors. They may be more or less suited to it, but they can fit. Consider the classic light cycles sort of game: it could be bricklayers madly working buiding walls, it could be snails leaving glistening trails, it could be ditchdiggers, it could even be something that is really a stretch, such as debaters competing in an argument. The mechanics are unchanged.
Now, our enjoyment of the game, and in fact the actual experience, may well be completely different. And what’s more, I’d vehemently agree with you when you say that some of those are bad metaphors for one reason or another. But the fact that they are not ideal doesn’t mean they cannot be applied. Many, many games have poor metaphors applied; consider that one of the things that really helped the mechanics of Amplitude was the change to the metaphor of a guitar — and thus Guitar Hero was born. By aligning the metaphor and the mechanic better, it even inspired new mechanics, and the two layers thus interacted with synergy, resulting in something new and bigger than the inspirations.
I think we’re simply using language slightly differently when you say that the healing game is “purely mechanical.” What I mean is that I am taking those four layers, and choosing to change only the metaphor and therefore the narrative as well (since the narrative is dependent on the metaphor). I am leaving the mechanics the same. And what you mean is that since I am leaving the mechanics the same, it’s mechanical. š So I think that’s just terminology.
I agree that the industry is moving towards emphasis on the metaphor, and has been. I disagree that this is necessarily inevitable. The more you depend on the metaphor, the less likely you are to change the mechanics — this is evident in the vast majority of titles developed today.
I also firmly believe that you cannot allow the mechanics to get overshadowed, because the area where innovation can truly occur is on the mechanics level. There’s plenty of new stuff to be done with a broader selection of metaphors (that was my point in presenting the healing game, after all) but to my mind, there’s also tremendous value in innovating on mechanics for their own sake. Mechanics can suggest metaphors just as much as metaphors can suggest mechanics.
I even argue that often, the underlying message that we want to convey may be one that mechanics convey, not the metaphor. As an example, I’d offer up the Watersnake game I have referenced in a few talks and that was mentioned in Smartbomb. That game is mechanically a simulation of blood sugar levels and diabetes, but the metaphor is very different. Ironically, that makes it more likely to convey the information about blood sugar management, because people come to understand it systemically, rather than just the metaphor — and the netaphor and narrative of blood sugar management is really rather dull and scary to read about.
In the end, I believe the medium of games to be at its core more about the mechanics than about the metaphor, precisely because metaphor without mechanics has other names altogether: movies, music, etc. Mechanics + metaphor is what this medium is.
I don’t think this is correct. The mechanic is indeed explored by the variations, but the variations are statistical in nature. They are not created by the theme. They are simply given names by the theme. Similarly, I don’t think the theme is necessarily created by the narrative; sometimes it’s the other way around. I tend to think any of these can give birth to any other one of these.
Raph says: “the variations are statistical in nature. They are not created by the theme. They are simply given names by the theme.”
I say: let’s replace the theme/metaphor in Tetris with a new idea. In this game of Tetris, you’re connecting legos.
Fundamentally, it’s the same game. But the fact that you’re using legos allows you to use other shapes outside of the normal five or six. You are “allowed” to use one-block blocks, or eight-block blocks, by the new theme: legos come in those sizes, so your blocks can, too. This changes the variations. This changes how people play the game. The game balance is also fundamentally changed by these new variations.
This is not an isolated example. In virtually all games, you gain or lose variations based on the theme you choose, and based on how the narrative guides the theme from moment to moment.
Raph says: “Similarly, I donāt think the theme is necessarily created by the narrative; sometimes itās the other way around. I tend to think any of these can give birth to any other one of these.”
I didn’t intend to imply anything else: they all reflect on each other.
I like this post and the comments on it. They’re thought-provoking.
It’s entirely possible to arrive at your Legos example without going through the step of “Legos.” In fact, many Tetris clones did. š They also made hex versions, triangle versions, and so on.
In the case of Tetris, who knows whether the limitation of four on the content blocks came about because of mechanics or because of theme… we’d have to ask Pajitnov.
However, I’d also point out that a Lego theme doesn’t necessarily mean you have to offer all those variations on blocks either. š
Of course. You can have a variation in variations without a theme. A theme simply guides and suggests, and puts it into bite-sized pieces for the players.
For example, players will probably find hex Tetris easier to chew if it’s clearly about bees. The theme has a solid, exceedingly simple reflection in the gameplay. In this case, even the mechanics have been altered to suit the theme (or visa-versa). š
Random thought:
For what it’s worth, whenever I’ve thought about this issue, particularly the separation of “story” from game, my brain recalls what I learned about newspaper articles: Who? What? When? Where? How? Why?
What – Is this the metaphor? Chris Crawfords claims that to design a game, you must be able to answer “What will the players do?” In WoW, players go out an kill monsters.
How – Is this the bare mechanics of the game? Players kill monsters by pressing buttons at the lowest level, and moving/controlling their character at a higher level.
Why – This is what the “story” provides. Bartle’s and Yee’s player models describe motiviations (why) that players enter the game with: “I am a killer: I want to get to level 60 in WoW because it allows me to beat up on other players.” Story defines motiviations provided by the game itself, “I want to finish Fable because that will let me rescue my (character’s) mother and sister, and take revenge on the bad guy who burned down my (character’s) village and ruined my (character’s) childhood.”
When and Where – Typical setting of the world, I assume. Is this related to “content”?
Who – Me? Or my character? I’m not sure where this fits in.
Having said that, you claim that a game cannot exist without the core mechanics, “How” in my terminology. I’d claim that “Why” is also necessary since people won’t play the game without a “Why”. Even if chess doesn’t have a story (game-given “Why”), chess games aren’t started without an external “why”, such as “I want to beat my opponent” or “I want be be world chess champion” or “I want to kill some time”.
You claim that “How” (mechanics) is required. Chris Crawford claims that “What” (metaphor) is required (as far as I understand his writing). As a non-guru, I claim that “Why” (motiviation for playing) is required. Maybe “when”, “where” and “who” are also required?
http://raccaldin36.livejournal.com/849808.html
The way I interpret his writing is thus…
You have a thing you’re trying to build. It has four customizable parts. One of these parts, called “mechanics”, is absolutely necessary. The other three may be included if you wish, but generally speaking, all three are included as if they were one.
Multidisciplinary is such an objectionable term. I don’t consider myself multidisciplinary; I merely study as much stuff as I can, admittedly with a strong bias towards a number of accepted fields of study. When you say you don’t need to go to writing school to become a writer, there arises the question of whether or not you can deliver a factually correct story about quantum physics, having never heard of it. The answer is, obviously, no.
Programming, I was taught, is much the same way. You can’t program a piece of software monitoring a patient’s heart condition if you don’t understand the electrical pulses and all that jazz; or so the textbook told me. =)
So it is, I’d argue, with game design. The core of game design is mechanical manipulation, that single module. It’s making stuff up like chess or poker or football. These are each totally mechanical games; the universe in each is strictly defined, even if each cover separate domains.
But say you take chess and you name the pieces King, Queen, Pawn, Knight, etc. Now you have content (or “statistical variation” as Raph calls it. *bleh*). You can produce the metaphor of kings being useless, but important, and the presentation that women are all-powerful. You can dictate a narrative, telling us that the armies of Light and Darkness have met upon a checkered battlefield to fight until a King is captured under the rules set down by Euclid, God of Squares.
You could. =P
What’s most interesting to me is that Raph is saying that you can begin with any one of these pieces and build a game from there. I agree with this.
A friend of mine is writing an RPG. He has plans for two sequels. He knows the entire storyline. The narrative is done. Sometime within the next month, he’ll produce a formal draft of what the combat mechanics look like so we can criticize it. That’s narrative first.
Raph has talked enough about “fun” translating to “lesson” that I’m not going to rehash it, suffice to say that you can begin with, “What do I want my game to teach?” I’ve been experimenting with this idea. That’s metaphor first.
I have no decent example for content or narrative first, but I do have a more holistic example.
Let’s imagine how Gygax or whoever it was came up with D&D. There are a lot of ways they could’ve, and it’s a complicated system, so I’ll just gloss over most of it. “Let’s make a game about groups of adventurers that go gallivanting across worlds, fighting monsters, saving the day, and being heroes.” (Narrative.) “Well, they’d have to have swords and magic spells and stuff.” (Content.) “Oh, how will they know what can do what? Well, we’ve got dice… they’re pretty random. And it’s unlikely you can roll a 20, so that’s our critical hit…” (Mechanics.) “And we know that small groups of people can cause big changes in the world, if they work together and do the right thing. Let’s write a supplement about building castles, running kingdoms, and such.” (Metaphor.)
What people have generally done is taken D&D’s core mechanics and changed the content and narrative, resulting in some metaphor shifts, but not many. Assuming that MUDs are descended from D&D, they realized that the mechanics are easiest to implement, because they were easily computed. Statistical variation is simple, obviously, especially with rand() functions.
I don’t think that the chess pieces are statistical variation… I would have to ponder it, but to me the chess pieces are part of the system mechanics. They are essentially abilities.
Statistical variation would be differing arrangements of those abilities. If the pieces had hit points, for example. Or a given chess problem, for another example.
In RPGs it’s obvious, it’s the stats on the opponent. In games of tennis, it is the “stats” that differing opponents bring to the table.
Hmm, I wonder if “content” isn’t just “the factors that differentiate one opponent from another” — where “opponents” can appear at the atomic level (in my grammar’s sense).
Ah, I was a bit worried someone would trip up on that.
The terms “King” or “Queen” is statistical variation. You can replace it with “General” and “Champion” without a change in the mechanics of the game. Depictions would also be statistical variations, I think.
The mechanics governing the placement and movement would remain unchanged.
Who? The player and the content (in the statistical sense).
What? Manipulate the system into the prescribed outcome.
When? Where? This is content (in the narrative sense); after all, it becomes a narrative as soon as there are multiple “whens.”
How? This is gameplay. It could be the narrative too, but I think it’s the gameplay.
Why? This is metaphor.
I don’t think the terms are statistical variation either. I think they are part of the narrative side of things. š They are the metaphorical names we give to bundles of abilities, so I would call them part of the narrative layer.
Mechanics: a grid-based system of force projection with tokens that project force along multiple dimensions; a symmetric game wherein the objective is the removal of one particular token of the opponent’s.
Statistics: different configurations of tokens.
Metaphor: a battle in a feudal system.
Narrative: kings, queens, pawns, etc.
[…] Raph has posted about layers of game elements. I have replied, he has replied, I have re-replied, and I suppose he’s likely to re-re-reply.There are four layers, according to Raph, and it seems to be a fine distinction:Mechanics, variations, theme, and narrative.These all reflect on each other, providing and restricting how the others can manifest. In a simple example, the nature of the gameplay in most FPS games means that melee weapons are essentially useless, which restricts your choices of theme and narrative. It can just as easily go in the opposite direction.The thing is, these are all one thing:Gameplay loops circling the mechanics.Mechanics are the basic rules of interaction and “logic”. You can shoot, you can run, you can stack bricks, whatever.Exactly what shooting, running, and stacking you’ll need to dance with is a question answered by his “variations” layer. Here, you might have to shoot lots of little, fast guys at long range. There, you might have to shotgun a melee demon.These variations provide the many methods of interacting with gameplay to keep things interesting. Sometimes, the rules are simple. In Tetris, the variation is “one of these six blocks will come next, at random.”I think this is intrinsically tied to “theme”. Your theme guides your gameplay variation so closely they might as well be thought of in the same breath. You don’t put much sword-themed into a ranged FPS, because the variation doesn’t exist. You put gun-themed things in, instead.But what drives these variations? When you get sick of “one of six random blocks”, how does the game keep your attention?Why, it puts variations into the variations. It leads the theme around by the nose, thus producing hundreds of new variations. As you go from level to level in a game, more than just the aesthetics change. Maybe one level is more open, allowing for more sniper work. Another has more puzzles. A third is about platforming. Any and all of these are gameplay variations which are “allowed” by the narrative. By which I mean, knowing the narrative makes the players think, “oh, okay, that makes sense” instead of “why am I surfing down the side of a pyramid?”There is often more than one kind of gameplay mechanic, and the various gameplay loops interact with the various mechanics in various ways. This is what keeps our attention. These variations.Here’s the real question: is there a layer above “narrative”? […]
Hm… your “narrative” then describes far more than I thought it did, and “content” far less. I think what threw me was where you said “statistical variations” was “monsters”. It might be more accurate (or at least more clear) to describe it as “classes”?
It would seem the names and the look of the pieces [in Chess] would not fall under Statistical variation, but under the Metaphorical layer with a hint of Narration. Statistical variation in Chess however, sprouts from the number of tokens and their starting positions. Playing first or second also changes the player experience.
There is a hierarchy here. The ranks relate to the final product, not to the development process. Each layer after the mechanical layer refines and adds meaning to the pervious layer(s).
This is very much related to the Onion Metaphor that is on the second oldest page of Daniel Cook’s blog at http://www.lostgarden.com. It’s a good read much like the rest of his site!
Quick thoughts:
“Let’s pretend”, the whole “Magical Mystery Tour” may be examples of games (loose definition) completely on the metaphoric level and without mechanics, nor variations.
I’m not stating this, actually. Just speculating.
Sort of, instead of four possible starting points, your model would have two. But still working as a hierarchy. With either the mechanics or the metaphor at the top.
I still see the two as connected. And my belief is that they will become more and more so as we refine the “game design” and move to have more “genuine” designers that aren’t just derived from other roles. When game design acquires legitimacy on its own. Its own independence instead of being just a patchwork of everything else.
I’m starting to believe that your idea about the mechanical level being the one really indispensabe (which I tend to agree) is like the metalinguistic with the language. A game with just mechanics is a game that looks to itself and speaks of itself. Autoreferential.
I see this happening, but I don’t see this as a good thing happening.
Simply put: in the maturity of the genre I (hope) to see the two levels (metaphor and mechanics) being more and more tightly connected.
They are unconnected and relatively independent now because the genre is still immature.
Then I also believe that the innovation on the mechanical level passes through the metaphoric level. But again because I still don’t (want to) see them as revolving independently.
Abalieno, you’re quickly becoming extremely interesting. =P Here, I kinda stole something from you, so feel free to steal it back or whatever. Public domain and fun-ness.
http://raccaldin36.livejournal.com/850058.html
The problem with conceiving of game design as its own separate field is that it’s dependent upon other fields of a far more atomic nature. Or rather, one: math. I’ve come to believe that math can describe damn near everything, so I guess it’s a bit of a moot point.
To take the example of my friend the RPG designer again, I would point out he’s designing everything holistically. (I would’ve called him on it by now if he wasn’t.) He might have come up with the overarching story first, but when he talks shop with me, we cover some of the more particular points of the narrative, like exactly why his magic system works and what justifications he can give and so on.
I don’t really understand what you mean by self-referencing games. With language, you talk about language using language. With math, you talk about math using math (see: Godel). But with games… games don’t describe things, the way language does, the way mathematics does.
Hm… or do they? That’s what you mean, I suppose… I think you just glossed over an extraordinarily fascinating philosophical concept… But I digress…
So, we have a game. A game, like anything else, is an amalgamation of lots of other stuff. To bring in a hopefully useful analogy, so is a book. A book’s mechanics, to make this quick, is to turn pages and read left to right, top to bottom. Now, a book is about something. It’s about something because of the sum of the language, which is also about something, is summarily about something bigger. Story, chemistry, how to fix your cabinet, whatever. There’s an overarching about-ness.
Now, a game has mechanics, but a game also has metaphor. A game is no different than a book, in that it can have the same mechanics (page-turning) with vastly differing metaphors. But the same metaphor (the same story, let’s say), can have different mechanics (a movie, perhaps, or a pop-up book, or a book with pictures, or a book in Chinese).
In each of these different instances, the book itself is inherently different. It might be a film reel, or a children’s book, etc. You can hold its metaphor as a Platonic Ideal separate from its expression through a medium, which might govern its mechanics.
But sometimes the metaphor demands the mechanics. If a book is meant to be read sequentially, it’d be ordered that way. If you’re supposed to read a random page after random page, it’d be a deck of cards that you’d shuffle. Vice versa, the mechanics demand a certain kind of metaphor. A deck of cards can’t express a specific, linear story. It might express hundreds of possible stories, some more sensical than others, but the story is neither necessarily specific nor linear; you might as well bind an order if it were. (Okay, analogy’s a bit weak.)
So… a game’s mechanics and its metaphor are likely intertwined, for the same reason some stories work as novels, but not as movies, and vice versa. But in other cases, it doesn’t matter. (I want to say Spiderman is a good example of this, but I’m very hesitant, since it’s a comic. However, I have read some actual novels, and they worked, IMO.)
Just some musings. I’d post it on my blog, but I have a 2-post-per-waking limit for myself currently. =P
Content in Chess is the other player. š
[…] Raph’s Website » A bit on how I think games work […]
Marble sculptures require a block of marble as a starting point. The block of marble is absolutely necessary: it defines what makes a marble sculpture a marble sculpture as opposed to some other thing.
The art-historical genre or aesthetic within which the sculptor operates is contingent. But that contingent choice may dictate how big a block the sculptor needs, or what kind of shapes it might need to contain. It might affect whether the sculptor needs other materials (e.g., has an aesthetic dictate to combine marble work in a mixed-media format).
The representational grammar and individual vision of the sculptor determines what will actually be made in the final process of sculpting.
What part is more important? The material predicate is absolutely necessary. The choice of that material conditions some of the expectations that viewers may have because of the history of art (we expect something different from marble sculpture than we do elephant dung and plastic). But what matters is the final content. If the artist wants to make a mixed-media installation that the audience experiences in a sensory and immersive fashion but only has a block of marble and a chisel, they’re pretty much out of luck.
I’m with Abalieno on this one, I think. You can build a game in any order, but confusing the necessity of a particular material or structural predicate with an inferred importance or centrality to the goal of game design is a pretty good nutshell summary of why a decent amount of contemporary gaming product goes wide of its mark. Houses need foundations, but a house reduced to its foundation is just a ruin. What matters in a game or a house is what is built on the foundation–and a foundation (or a mechanics) forms a kind of limit or constraint on what can be built on it.
If you start by saying, “Let’s talk about a Healing Game”, and act as if that’s primarily a re-skin of mechanics, I think there’s something really fundamentally wrong from that point on.
I’m a bit in the middle here. The outcome depends on where you started.
Some have used the Foundation>House example, but that’s more about the structural approach to execution. At the conceptual level, what a structure looks like is generally defined by both aesthetic and functional goals. Of course, Louis Sullivan comes to mind. Yet, technologically, we’re long past the point where form absolutely must follow function. I’ve always come down more on the side of Charles Eames in these regards. Form and function are a partnership, because aeshetic and usage or symbiotic.
Metaphors and mechanics can be replaced, but not without consequence. A poorly-themed well-playing game can do ok, but probably not pull down the numbers of a well-themed well-playing one.
This is why I agree with the “multidisciplinary” point about game design. It’s because it’s part of all design. Swiss Army knives as it were. Graphic Designers do not need to be print specialists. Industrial Designers do not need to know in complete details the fundamentals of plastic flow into tools. Game Designers do not need to be expert story writers. The trouble is, “Design” is fungible, a term so abused, it lacks clear definition.
To me, based on my experience only (and therefore, of course, heavily biased), I consider “Design” a bit differently. Since the conceptual work on any project is usually the least of the effort, I consider “Design” to be a bit more akin to project management. However, as can easily happen, too much project management can result in a loss of the original vision. A zillion little compromises later and suddenly the original concept is no longer recognizable.
Therefore, I’ve had to come up with a new term for it:
“Vision Management”.
We’re charged with ensuring the vision we’re paid to be experts in is executed across the multiple disciplines. Jack of all trades, masters of none, hard to quantify, but easy to qualify. Sometimes we need to be conversant in certain areas. Other times we need to be able to manage relationships. It’s a bit more of a mix of that adage: “the science of information management, the art of relationship management”.
What do you think?
This reminds me a little (and only a little) of a piece I wrote on my own web site, from the point of view of a player looking at MMORPGs and the attempt by most developers/administrators to try to “balance” them. I suggested that a game consists of its rule set (which can be balanced), in-game characters (which can somewhat be balanced), content (which can be balanced), and the human player (which can’t). I’ve seen MMOs try to balance things such that no player can outperform another player regardless of the character he creates and how he plays it. If you try to balance to that level, you’ll never get there. Some players are better at manipulating your rule set than others, some have more time to play than others, etc. If your goal in “balancing” a game is to make it so no player has an advantage of any kind over any other player, then you’re chasing a mythical balance that you’ll never achieve. If somehow you do manage it, you’re essentially reducing the game to tic-tac-toe. If you want to see my whole diatribe, you’re welcome to read the article on my blog at: http://mikesalsbury.com/mambo/content/view/388/.
Here’s the question – is there some level where the fun-as-defined-by-Raph happens, or can it be spread equally across the four facets of the game? If so, how? If not, why not?
–GF
Just some quick provocations here. But before I start, let me just say Wow, what an incredibly rich thread. š Thank you!
Can we come up with games that clearly consist or depend 90% on one given level? Just an an exercise?
Hmm, hierarchy means one of them has to be on top; we running into a language barrier here?
In any case, the provocation: can we come up with four game concept descriptions for the same game (we could try the healing one, or something else), one for each of the four possible starting points? I think it can be done.
What is the name for the “mechanics creation” part? Because I would tend to call that “game design.” I would tend to call the other “interactive entertainment design.”
Say, rather, formalist. The reason I say it is indispensable is because I don’t know of any games without rules. (I have argued before that many of the games we consider to not have rules actually have many unspoken ones). Therefore, the process of putting rules into a game is, of the four, the one that makes the construct into “a game.” Can you think of examples to the contrary?
I don’t see this happening; where do you see it happening?
A provocation: draw an analogy to any other medium where it has worked this way; I think we’ll tend to find it’s exactly the opposite?
Must innovation pass through only one of the levels? How would, say, Abalone have gone through the metaphoric level? Perhaps Quoridor did. Blokus?
But they do describe something, arguably math. They model, don’t they?
What about a game like Fluxx?
Modern emphasis on form has given us books like Cortazar’s Hopscotch (Rayuela in the original), Robbe-Grillet’s book-in-a-box of loose pages and so on. Are these alternate mechanics, and do they carry meaning?
Isn’t it possible to say they are always intertwined, but you can provide much greater emphasis to one layer or another?
Bronze sculpture requires no bronze whatsoever as a starting point. I don’t think you usually start on the marble block, either. š
More to the point, though — does that mean that checkers could not have been designed within a computer? That tic-tac-toe couldn’t have been designed with stones and cardboard boxes? (That the maquette of David might well have had as much artistry as the final statue?) Isn’t there some point at which we have to say that the medium matters, but there’s something that is core before the medium itself?
From the audience’s point of view, yes. What other points of view are there?
Doesn’t your very metaphor imply a hierarchy? I think here I’m actually more with Abalieno than you are: I think you can start on any of the four, and let that one provide the constraints to the others.
Remember: this was a thought experiment, not my actual process.
That said, the provocation I would offer is, “I think that this is what happens more often than not, and I agree that it’s bad.”
Darniaq: whoo hoo, Eames and Sullivan. š I gather you work in the design field. š I didn’t come up with any provocations for you. š
Mike: I’ll have to follow the link later when I have time. š
Glazius: I define fun as the learning of patterns… that can happen at either the systemic or narrative levels, I think. Not the metaphoric, and I think the statistical just helps perceive the underlying pattern.
I would imagine there was a time (or ever are times), within certain genres of music writing and presentation where the mechanics of making music and the theme, or experience, of hearing that music correlate heavily. At least its a sliding scale of sorts and however the effect is desireable or not is down to taste.
Its a stupid comparison tho, and requires the music to be analyzed through a nasty little filter. Its also not the opposite, just a minor note to say this is a somewhat subjective matter.
There is a great difference with music as you can be a happy music consumer without understanding the music, but you’ll have a hard time being happy about consuming a game you dont understand.
I really appreciated your approach to describing the different aspects of a game.
I also appreciate MikeRozak’s common-word approach to describing the same
I’d propose a variation on this. Each is not a perfect match for your classification.
When / Where?: Narrative & Metaphor
What?: Metaphor, Mechanics & Statistical Variation
How?: Mechanics & Metaphor
If your classifications are the dimensions of game design, perhaps the W words are the planes thus formed?
The two others, are different beasts, outside of your classifications.
Who: To my mind, ‘Who’ is/are the players. The ‘statistical content’, even when AI driven is not a player … it is an advanced form of the ‘How’ or core game mechanic & statistical variation. To be a ‘who’ you must be able to experience the other W’s. Admittedly, this is a departure from the typical newpaper usage of the W’s.
‘Who’ gives you clues about the dimensions of a design … what mechanics to employ (perhaps older gamers prefer non-twitch activities, western cultures might have preference for common western narratives, metaphors, and even mechanics) As you’ve pointed out, eastern games tend to be a bit more direct with with the PvE.
To consider AI driven mobs as part of the ‘Who’ is to suggest that a game’s core mechanics should be design to suit Lord Vader’s enjoyment.
‘Who’ is strongly related to ‘Why’. I would suggest that ‘Who / Why’ be thought of as the fourth dimension of gameplay. A dimension perpendicular to all others.
For each of mechanics & statistical variation / metaphor / narrative the question needs to be asked, ‘Why is it enjoyable for my who?’. Why are the mechanics enjoyable. Why is the metaphor attractive? Why is the narrative engaging.
A design that merely posits the first three dimensions in a consistent manner is a potential game. A design that links each of the three dimensions with ‘why is it fun?’ is potentially good game.
I hope your new book, Raph, will link ‘How Games Work’ with the ‘Why?’
I’m kicking myself because I can’t remember the name exactly, but it was a multiplayer game, played face-to-face. The only things in the game are the rules, and a set of meta-rules which provide a format for changing those rules.
The object of the game is to change the rules of the game so you win.
–GF
Are we excluding “degenerate games” like “Mornington Crescent”, or are those perfect answers?
Mechanics and statistical variation seem pretty entwined to me so much so that I wonder if statistical variation isn’t just a type of mechanic. In chess and Go it certainly is. The only variation comes from choosing how to move. “Knight”, “Castle”, “Rook” isn’t variation certainly — it is a choice of metaphor. Changing the layout of the board might be considered a change in “content” but it is first and foremost a change in mechanics. Thus I think that statistical variation is inevitably just part of mechanics. The list of rules for your games just gets quite large once you introduce complex content.
Metaphor, narrative and presentation also seem entwined. It’s not clear to me why metaphor or presentation aren’t simply very simple instances of narrative.
So really I think this is just a way of restating the traditional ludological/narrative dichotomy. You have “content” which is a type of mechanic and you have metaphors and presentation which are simple types of narratives.
Coming from around eleven years of strategic branding and marketing communications management experience with a background in graphic design, I strongly disagree. Graphic design is an applied field in which practitioners develop material for the purpose of commercial communication. Tradition holds that graphic designers create printed commercial works. We have other classifications for those who do not work primarily with print, such as “graphic artists” and “web designers”.
Professional graphic designers are print specialists; however, they may not be experts in printing. Since our core medium is paper (e.g., business cards, brochures), we need to possess extensive yet practical knowledge of paper (e.g., stock, weight; “#120 Classic Crest White – Cover”) and associated production processes (e.g., binding, typesetting.) Graphic designers generally communicate using sight and touch — two of the five senses. Some clever works also communicate using sound, smell, and taste. Also recognize that paper is not the only print medium. Graphic designers can use metals and plastics too.
The following describes how I segregate master designers and everyone else.
The purpose of design is to provide an effective solution to a problem. A master graphic designer uses at least sight and touch to develop effective commercial communications. A master graphic designer can also adapt the design process to the situation to develop effective commercial communications. For example, graphic design for interactive materials (also called information design) may necessitate sound to satisfy principles of user interface design. Graphic design for fragrances may necessitate smell.
A master game designer uses at least a single layer to create an interactive experience; however, a master game designer can also adapt the process to the situation to create an effective interactive experience. For example, a master game designer would be capable of designing an interactive experience, starting with any of the layers Raph described.
Remember: a hammer is not an elixir. A hammer cannot be used to solve all problems, but a hammer can be used effectively to solve some problems. List several problems in game design and then think about which layers would effectively solve which problems. That’s a creative exercise that separates the master designer from everyone else.
Did you ever open a topic to discuss your definition?
I’d agree that they are entwined concepts. To me, the purpose in separating them would be to enhance the process of design. Lumping them together, and you might get great missions but poor story arcs. Separate them in the process, and perhaps better focus can be given to both. Aren’t all big tasks best achieved by breaking them into smaller, more easily achieved components?
Sounds ike Nomic or Fluxx. Nomic is more thought-experiment than game, and Fluxx is a card game where win conditions are mutable.
Great question. I agree they are very entwined; at the same time, I have trouble seeing an orc versus a kobold (or rather, “a level 3 entity with x hit points, etc etc,” and a “level 5 entity with y hitpoints, etc etc” as being “changes in rules.” YMMV on this one for sure, because I don’t have a bright line.
The way I had always thought of it was that content used a fixed set of rules, but put through minor variations; a true rule would affect the verbs a player can make use of. The verbs in dealing with the different levels of creature don’t really change.
But none of the abilities of the tokens change; only the topology of a given situation. It’s comparable to a boss in a 2d shoot ’em up: the patterns of where the bullets fly might be different, but your verbs are still limited to moving and firing, and your lose condition is still getting hit. It’s a different problem to solve with the same tools — that’s what I mean by statistical variation.
Another very good question. I suppose they intuitively feel different to me, though again very entwined. We can have many many narratives that use the same metaphor; the same narrative using different metaphors is far more difficult, I think… though it’s probably worth thinking about where stuff from narratological theory fits here. I mean, is the Hero’s Journey a metaphor or a narrative? I am unsure.
That said, I do think it’s easy to see games that make use of metaphor but not of narrative. Keep in mind that the narrative layer, as I describe it, is really almost a presentation layer — it includes all sorts of other things. Hmm, I need to think about this more, but you can apply the metaphor of a battle to chess, but then whether the narrative is LOTR pieces or standard is relatively minor.
No, but there’s been discussion scattered throughout some earlier posts & on various sites that have reviewed the book (Grand Text Auto, for example).
As far as what you start with, mechanics or narrative, I think you can start from either direction although inevitably you must give good consideration to game mechanics which I do consider to be somewhat primary. You can start with Tetris LEGO’s as a design idea, but if you can’t work out effective game mechanics for this then you won’t have a game. You can express any story you want, for example, Tetris with 5 dimensional beings who bicker incessantly about their favorite 5-d TV shows. If you doesn’t make sense mechanically, however, then you don’t have a game.
I think another distinction needs to be made about narrative, one that cuts along different boundaries and addresses a narrative’s potential dependency on game mechanics. Consider the following set of game stories as they progress from purely narrative to purely mechanical:
1) In a climactic finale, Cliff, the protaganist is betrayed by his lover Samantha, however Jane, who we thought was dead, returns and saves the day. (note that this is not a game)
2) A scene is shown with Samantha attacking Cliff. The player takes control of Jane and must successfuly enter a sequence of keys in order to successfully distract Samantha. Control shifts to Cliff and the player must successfully aim and shoot Cliff’s gun to disable Samantha. (a game but with a scripted story, the player’s actions lead only to success or failure of the scripted story).
3) Cliff is attacked on a battlefield by Samantha and at low hit points, but Jane casts “stun” on her. He recovers and attacks with a “power attack” that kills Jane. (less story-like but with almost complete player control)
Some deem the earlier choices as more meaningful but I think there is an argument to be made that the latter is more meaningful as it is the result not of a script but of actual player choices. Some might argue that the metaphor of stunning and attacking makes the last scenario still narrative. A lot of ludologist/narrative arguments might be had here: is this mechanics or narrative? It’s probably both. I think that it is mechanics causing story. It is completely mechanics driven (the mechanics of certain “specials”, hit points and the mechanics of a battlefield) but also has substantial story (we can imagine players recounting the story of such a battle, and there may be player-created politics to the encounter as these 3 may have a history). In our own world all meaning is derived, inevitably, from mechanics. Talk all you will of bravery, love, honor, betrayal, etc., but none of these concepts would exist if we weren’t living, breathing, bleeding, beings acting a world defined by physics.
So I am arguing for a dichotomy of narrative other than metaphor/narrative. I think it is important to cut away artificial narrative (that which has been scripted and inserted into a game) versus emergent narrative (that which emerges from mechanics). Personally I think that emergent narrative is where we are headed with our MMORPG’s and I think that it makes the ludological/narrative schism moot. Either we say that mechanics are primary for such stories or we say that our narrative are themselves heavily invested and dependent on mechanics.
Is it… http://en.wikipedia.org/wiki/Nomic ? =)
I need to sit on this. I put that down shortly before I went to bed last night, and I only recently woke up, so I haven’t actually thought about it. I’ll probably have something tonight.
Straightforward, of course they are alternate mechanics. With Hopscotch (which I had to rely on Google on), I’d hazard to say that the mechanics are primarily still page-turning and page-reading, but there’s the additional mechanic of “Should I read the last 99 chapters?” and “Which chapter should I go to next?” I think it’s interesting that these are questions, rather than instructions.
Perhaps, instead, it is statistical variation? That fits neatly into the hierarchial idea…, naturally.
Hm… the more I think about it, you’re right. A game without its metaphor is like a mathematical system that doesn’t describe anything, which is precisely what Godel said wasn’t possible. Does that analogy actually carry over? It very likely does. But I have to push this line in my head harder.
Adding a layer of Narration to Chess would be like giving the pieces names like King Alfred, and pawns Bob and Alexander! Even some scripting too, like forced moves at certain points. š
However the mechanics themselves are built in order to make the boss battle fun in both situations. I.e. the rules for how one gets hit, how often one gets hit, all the “core” mechanics depend on the mechanics of the “content” and not just the other way around. So much so that I think that content really is just mechanics a lot of the time.
Viewed from a game theoretic standpoint, the mechanics really should be all the possible game states and the rules for transitioning between game states. Content is a required element of this. Change the starting positions of the pieces in chess and you have completely changed the state space. It’s a different game, with different mechanics (requiring new “opening books”, strategies, etc.).
StGabe, on your earlier post, I think we are (for once) nearly totally aligned! I’d go further and say that your example 2 is actually a story with some embedded games — it isn’t even the narrative layer I am talking about. (Just as a game that is all game but intersperses cutscenes is a game with embedded movies). At that point, it’s mixed media.
On your second point, though, I guess I still feel like there is a fruitful distinction to be made between the game of chess, and considering each chess problem to be a unique game. I accept that the boundaries are very blurry, but it still feels like these are different things. It’s like the game is a function and the content is the inputs, like the game is the engine and the content is the fuel.
Ketzup wrote:
More importantly, IMHO, adding a layer of narration to chess gives the players a reason to like King Alfred, BoB, and Alexander, and a reason to dislike the enemy king, queen, etc. from the enemy’s side. Because the player likes the pieces on their side (on a personal level), and dislikes the enemy’s pieces (on a personal level), they will become more emotionally involved in the game.
Adventure games and most CRPGs use this technique. Chess doesn’t. Instead, the sole reason why people play chess is to beat their opponent, exercise their brain, etc. They don’t play because their queen has a nice personality.
I think both of you need to stop trying to build a game out of a mechanic. Games built out of a mechanic are a gimmick. A gimmick can be a great game, but lacks both in depth and long term appeal. And as both you have proved a gimmick doesn’t have to be simple.
So a debate about a game mechanic… even I’ve fallen into doing such things. I’ve come to realize it really doesn’t change anything. You’re providing a means to achieve a goal. If you can make it fun then you have a start. If you can make it fun, but challenging you are heading somewhere. If all you can do is debate it then you are getting everyone no where.
But thats why people like myself and Abalieno are around… because hell we just like talking. People like Raph are around because he got a job talking about such things.
This whole arguement stinks of Kosterisms (no offense intended), vaguely specific terminology, and throw away scenarios. I think it was said best… the “healing game” is already made and its name is Trauma Center, but it just hasn’t been put massively online with 1,000s of other players.
Chess and Tetris are gimmicks?
No they are not, but you are both thinking way to in depth and getting lost while you are there. You both are stating a game mechanic and then abstractly pointing to other games and going “Look what happens when you change the rules of XXXX” and then it is “You can’t change the rules of XXXX”. Changing the name of chess pieces does nothing to effect the outcome of the game? Correct?
What started as a mechanic, a way to play the game, has turned into a rule of the game. That is a line you have to draw. Rules and boundaries are one thing… mechanics to play within them is another.
The whole conversation ended at…
Is it light cycles or brick layers that float your boat?
I haven’t seen Puzzle Pirates mentioned in this thread, but here’s a game that takes very basic games, tetris-clones, etc, and applies a very thin metaphor. Basically, you play tetris and they call it sailing, you play breakout (or whatever) and they call it working the bilge, but the raw mechanics are right there on the screen for you. MMORPG combat and crafting systems could stand to take a lesson in terms of giving players a way to engage in the combat and crafting metaphors without resorting to simply the bars and buttons mechanics we are all used to.
I mentioned this in another thread here, but imagine embedding something like texas hold-em as the core mechanics under a fantasy combat metaphor. You know your starting offensive and defensive states (pocket cards), you don’t know your opponent’s. You commit as heavily to the combat as you wish from a pool of your own power (chips), and only this is what your opponent knows about you. You can retreat from combat at any time, taking damage only equal to what you have already committed yourself to: did you charge into the fight full-steam ahead (raise before the flop) or hold back with your shield up, ready for an onslaught (limped in for the minimum)? Then common, shared states and environments are revealed; is it muddy, is there a slope, is the sun behind the defender (the flop comes down), and again the players can choose to commit to combat even more heavily, up to their level of XP, or HP or whatever (up to how many chips are in their stack) or to retreat. If neither player retreats, the combat is resolved and the winner gains power while the loser takes damage, but is not necessarily killed (unless he’s gone “all in”).
It’s interesting to see how frequently the metaphor and narratives are varied but how the fundemental mechanics seem very stuck in “bars and buttons”.
One of the problems is that the boundaries tend to be really fuzzy. The chess example is a good one: the mechanics are about projecting force and controlling territory on the board, I think most will agree. But, are the pieces part of the mechanics, or part of the “statistical variation”? I’d consider them part of the statistical variation, because one could introduce a new type of piece with a unique movement (say, move any 4 spaces) and it “feels” right in the game. The issue here is that we’re used to one specific set pieces and moves, so it is easy to think of them as core mechanics.
So, given the fact that variations are fairly rare in chess, does it really make sense to talk about the game as if you could invent new pieces? Or, is it more useful to talk about the game as a whole, considering the established set of rules as “mechanics”, and investigate it from that angle? Where is the line really drawn?
My input,
The distinction between mechanics and content depends on context. Like I said, game theoretically I think that content is absolutely a mechanic. 99% of the theoretical understanding of chess relies completely on the starting board. I do agree though that content can be seen as somewhat different but then I also think that where that line is drawn is somewhat arbitrary. If the starting positions in chess are content and not mechanics then why aren’t the move values of the knight? You can think of the movement of the knight as a general mechanic with parameters (2,1) just as a kobold might have 1d4 hit points.
It comes down to whether the distinction is useful or not and I think it can be although it personally makes more sense to call content a specific type of mechanic, distinct from more pure mechanics but still a rule of the game, with a certain form. It’s just a rule that (like a lot of rules actually) can be a parameter of another rule. I particularly think so in a discussion that I see as still largely about ludology versus narrative. Ludologically speaking, content, at least the content that exists to create statistical variance, is absolutely a mechanic, or gameplay or whatever you wish to call it.
Now metaphor is not narrative if by narrative you mean something that has a scripted introduction, complication, climax, etc., and I can see a useful distinction there too. I do think that metaphor is often directly the form of emergent narrative. In fact, I think that as emergent narrative more strongly dominates artificial narrative, the artificial elements of the story tends to be expressed through metaphor rather than plotting. If we do take your 4 categories as a heirarchy then it seems that emergent narrative is a movement of story down the heirarchy with the story dropping traditional narrative forms and instead narrating through metaphor and mechanics (or rather, people engaged in the activities proscribed by the mechanics).
Heartless, I have to admit you kinda lost me. To me, the light cycle game and the bricklayer game will feel really, really different, even if they might have basically the same mechanics. Perhaps a better example, because it’s both real and also not exactly the same game, is stuff like Jumpman vs Pac-Man vs Miner 2049er (I realize these are very very old-school references… sorry). Or perhaps UT versus Quake.
Baj, it might even be more fruitful to think of Puzzle Pirates as keeping highly similar narratives and metaphors as any other MMORPG, but swapping out the mechanics. That’s certainly the layperson’s way of describing the game: “it’s just like any other MMO except the combat is like Puzzle Fighter…”
Brian, StGabe: I see what you are both saying. But to me a rule implies a verb (cf the grammar stuff) and the statistics do not do that. They are data verbs act upon, but not verbs themselves. That’s the distinction I am trying to make.
I must confess, I’m not sure how more dynamic content enters into this. Player housing certainly doesn’t seem to be a rule. It is the result of a rule or mechanic however. Really I just want to maintain, linguistically, the tightly coupled nature of mechanics and content.
Raph: So, I’ll grant that perhaps the fact that there are pieces and that they move could be considered a core mechanic supporting the “project force and control territory” basic mechanic. But, as StGabe pointed out, aren’t the mechanics of a knight’s move just a variation? You could just as easily say the knight can’t “jump” over other pieces and make it more like every other piece. Of course, this might be my own personal bias since a majority of my practical work has been focused on the “statistical variation” level of design in modifying (sometimes to a large extent) an existing game.
My original point remains, though: it can be hard to tell where the line is drawn, exactly. Likewise, I think the line between metaphor and narrative can be a bit tricky to nail down. Precision in these terms will help us talk about things more intelligently.
There’s also the issue of related design issues like user interface design. I’m sure you’ll agree this is an important part of game design, but where does it fit within these four levels? It doesn’t fit neatly within a single portion.
My further thoughts.
I don’t think that the movement of a knight is actually a statistical variation; the game of chess does not admit, by its formal rule construction, variant movements. Because of that, to me a knight is actually an ability. Changing how it moves is actually a rule change.
Again, I’ll define systemic content (just to leave out the whole narrative side of things for a sec) as “stuff that changes in the game mechanics that doesn’t change any rules.” The length of the pole in the pole vault, and how high you have to clear. The stats on the orc versus the kobold. The varying capabilities of the other players in a tennis tournament. The difference between golf courses. The differences between one level and another in a platformer.
I would bet money that one defining characteristic of content versus core system mechanic is that a grammar will encompass everything in the mechanic, and nothing in the content.
Another bright dividing line might be that the content is always expressible as tabular numerical data. Changing the way the knight moves isn’t — its ability to “jump” is not quantifiable, and you would have to resort to adding a boolean column to the data.
There are some reasons why content differs heavily from statistical variation and mechanics, the most obvious one in alot of games is that content carries alot of motivational values.
A high level player in WoW will consider a 74 dps weapon to be content, while a lvl 4 robe is statistical variation š (Altho technically they are not different, only their usefulness varies alot.)
In a game of chess the content comes from the opposition. There are cases where the opposing player in chess gets reduced to statistical variation, this happens when the chess computer is too sucky to be fun. There are variations on chess where you play across several boards with several pieces and pieces rotate clockwise around the whole table when they are taken out of play on one board (or something similar like that). Good content in chess is found in chess clubs with a ranking suitable to your playing. To some players the mechanics themselves become motivation to play, and hence valued as content to these players.
In some games like Tetris the mechanics proivde all the requirements of the content through the scoring system, you will however find that mechanical games have a drastically limited lifespan in /played per user unless they have additional content added through online connections or similar networking of the scoring system.
In single-player Tetris the content is yourself. You will be motivated by the game because your brain rewards you for optimizing the processing of gameplay until you reach the highest possible tetris optima. From there on you can be motivated to beat your own highscore until you get one that is so good you eventually give up on beating your own highscore. Some games do this neural optimation relatively satisfactory, altho in the long run it really dosnt matter as anyone who cares about going through this optimization will reach a relatively level playing field. The setting and additional motivation can expand the enjoyment of gameplay beyond the optimized state, so it does make sense to skip alot of work and chose the rpg sollution where you basically avoid the phenomenon from the beginning.
Raph wrote:
I donāt think that the movement of a knight is actually a statistical variation; the game of chess does not admit, by its formal rule construction, variant movements. Because of that, to me a knight is actually an ability. Changing how it moves is actually a rule change.
But, does it change the the game to something different? You say in the post that the mechanics are the irreducible part that makes the game a game. You could change how a knight movies and you’d still be playing chess, although it would be called a variant of chess.
Let’s take an example from another classic and established game: poker. The official rules for poker don’t allow for wild cards, but home games do allow for wild cards and people still call the game they play “poker”. So, let’s think about your post in terms of poker. Consider the following comparison between variants of poker:
5 card draw vs. 5 card draw with wild cards
5 card draw vs. Texas Hold ‘Em
5 card draw vs. 7 card stud
Now, how do you classify these different types of games? Are they all the same game on the mechanical level, but with different statistical variations (hand size, ability to draw, community cards, etc)? Or, are they different games that happen to share some common mechanics?
According to the way I’m interpreting your definitions, these are all the same games with the following qualities (forgive errors, please):
Mechanics:
Collect a hand of cards.
Use your hand (and optional community cards) to make a 5 card combination.
Strive for the highest 5 card combination compared to your opponents.
Alternatively: force your opponents to concede (fold).
If you took out any of these basic mechanics, you would no longer be playing poker.
Variations:
Hand size.
Number of cards from hand required to be used.
Discarding and/or drawing cards. (Draw vs. Stud.)
Number of community cards.
Number of community cards required to be used.
Which players know about which cards. (Face up, face down, or “Indian”.)
Wild cards.
High/Low rules.
Betting rules: ante, blinds, raise limits, check raises, etc.
Opponents.
etc.
These can, and often will change between games. This is one of the things that make poker a very popular game, in my opinion.
I think the chess example is messing up the discussion, because the rules are very, VERY well established. But, way back when someone was originally designing chess, I’m sure they went through a lot of variations in piece movement, placement, and distribution before arriving at the game we know and love today. The special ability of the knight to jump over other pieces while moving was likely added later to make the piece much more useful.
Another bright dividing line might be that the content is always expressible as tabular numerical data. Changing the way the knight moves isnāt ā its ability to ājumpā is not quantifiable, and you would have to resort to adding a boolean column to the data.
This would be similar to adding in a “Flying?” boolean column for units in an RTS. I think this is perfectly valid. Now, I’ll agree that the concept of “flying” is a mechanic, but assigning that ability to a unit/piece is a variation. All in my opinion, of course. If you disagree with this assessment, I think you really need to go back through and tighten your definitions to clear things up.
My thoughts,
[…] Comments […]
Brian, excellent points on poker. I am going to have to think about that one.
One possible way to approach it would be to say “those are all different games in the same family,” in the way that violins, violas, and cellos are “mostly the same.” But I am not sure about that; violins and cellos are physically different, and in many ways I see the rules as being the “physics” of a game.
Another way to think of it is that the “physics universe” of poker allows for some options to be set, almost in a manner akin to how we set game options in a videogame. Once set, the rules of the mechanics are consistent, and the statistical variation content comes from other players, as it does in all head-to-head games.
Something I’ve always found interesting: the GTA series began life purely as a city simulator (according to the designer). Perfect example of a game which started from the mechanistic viewpoint.
Beyond Good and Evil however (and this point is pure speculation on my part, as I’ve read nothing by the dev’s of this game) seems to me to be very much started from the narrative point of view.
The interesting thing is that both games kind of converge to the same kind of level; open ended gameplay in a large world with a story to guide you through it. Both are great games, too. Seems to me that where you start from isn’t that important; you gotta start somewhere, and any starting point might get you to roughly the same ending.
Relating to the current thread of discussion:
I’m still not sold that the ‘mechanical layer’ and the ‘statistical variation level’ are really two separate ‘layers’ at all. The statistical variation level seems to me to be a way to analyze the difference between the same family of games. A way to explain house rules and a way to bunch up the poker variations in to the same game, instead of the same ‘family’ of games.
As Raph stated above, and I agree with, “the game of chess does not admit, by its formal rule construction, variant movements.” Exactly. However, if you were the one ‘playtesting’ chess, there would have been a statistical variants, however over time the rules have been formalized and have become immutable.
If you continue to apply statistical variation as a complete layer, then Mario is still the same game whether or not it contains the last level, if it contains the fire flower, or even how high Mario can jump. I would argue that the removal of any of those creates a completely different game. Certainly, it would still exist in the same ‘family’ of Mario games, however it would be a different play experience and thusly a different game system entirely.
To summarize, I don’t believe you can easily separate mechanics and statistics, as they are tied together too strongly by their nature as the mathematical constructs of a game(as Raph states, the ‘physics’ of the game).
Now, I could certainly be missing something key about the reason to separate out ‘statistical’ variation, and I will admit my thinking about this problem is largely influenced by Salen and Zimmerman’s three prong division of game structure (rules, play, and culture). In my mind, applying their ideas, that which is statistical variation in Raph’s model falls in to two catagories in theirs, both rules (for that which is purely mathematical in nature, ie. rule variations) and play (for that which is influenced by the player, ie. choosing which ruleset to use).
MacD, a further additional point on GTA: David Jones claims that the specific direct gameplay inspiration was Pac-Man. “The cops are ghosts, you’re Pac-Man, and there’s power-ups…”
An example of applying some mechanics from a different game into a vastly different metaphor, then evolving it radically.
Maybe we are all lost beyond hope?
Now lets take chess. Isn’t movement the mechanic so therefore any type of movement is the statistical variance layer added to the mechanic. If you change one you’ve now changed the other.
Now the rules define the movement for each piece. So your rules define your mechanic. Rules can change to create a different game. While it may be 5 card draw or Texas Hold’em it is still poker. The mechanics define that its still poker.
So the mechanic builds the game and is opposite of what I said before. We can conclude that Heartless is in the loop.
Nor do the rules of chess allow by their formal rules for pawns to start in the backrow. That this might be “content” and the movement of knights be “mechanics” is really an arbitrary distinction. I understand why you might want to make such a distinction for some cases but I still think that mostly content is a type of mechanic and I definitely think that the starting placements are just as important as the movement of the knight when understanding Chess theory. Both changes are just changes to the state space. One is a change to the initial state and the other is a change to the state transition.
In your pole vault example there is actually a rule as to how the bar is raised and there are rules as to what types of poles you can use.
Ahh, but content usually is expressed by a grammar. As indicated, the pole vault has rules for both the pole length and raising the bar. The starting position in chess is a grammar, just a simple one. In fact, as traditionally defined, grammars have starting strings and transition rules. The starting position of chess is the starting string of the rest of the grammar and is required for any formal analysis of the game.
Sure it is. Like I said, it is simply an instance of a function with (2,1) as parameters. The function is: move X in a direction then 1 in an orthogonal direction, with jumping allowed. More generally it wouldn’t be that hard to create a grammar that was a superset of all chess moves and then assign values to each piece to define their particular instances of movement (just as we would define a grammar for possible content in an MMORPG and then fill in HP, attack, etc., values).
[…] over on raph’s site, he talked about making a healing game. a gentleman over on the cesspit has taken up the flag in argument against it. while his argument reflects some of the same feelings i have, i don’t think he’s conveyed it very well. […]
Ah, but in these terms, we have to figure out anew what “game” is. I think we could just as easily argue that the bare mechanical form of the game is the jumping and the traversing and exploring. The “Mario” part of it is not the issue. See the new post showing the Tin’s Fruit Stand development path, which I am sure Abalieno will hate. š
OK, let me take another try at this because I think this is a valid point too.
The rules of chess define the machine. Prior to a game, there’s effectively a negotiation period wherein the rules might get altered — in some games it’s made very explicit, as in Brian’s poker example. In other games, it’s somewhat implicit or skipped altogether. In the chess case, the decision to work a chess problem versus play against someone is presumably part of this period.
The end result, though, is that we enter the game proper with fixed rules. The machine might begin in different initial states, though.
Now, in the grammar I define all games as being effectively head-to-head versus an opponent; the nature of the opponent varies wildly (the clock, your own body, someone else’s running of the same race, the computer, another player). You take actions with the system, and the opponent takes actions with the system that present you with a challenge.
By these lights, you’re right; the opponent is effectively just manipulating rules. There is no “content.”
But another way to look at it is that the opponent is the “content.”
I suppose that in this case, the pole vault example becomes more complex. The opponent in the case of a pole vault is effectively presenting a challenge that increments, but part of the opponent is also your own body.
Now, you state
Sure, and there’s also a rule for how much physical force can be exerted by a given muscle under given conditions, but it’s one we don’t really fully understand because it’s largely unstated (“imported” to use my term in speaking of paidia).
Of course; in any grammar, the content has to fit within the grammar, but it can be tremendously variable nonetheless. In this case, you have to define the rule of “jumping” in order to permit the content… hmm, I have to think about this one more.
This is a really fruitful discussion for me, btw. Lots of interesting challenges and examples. But at the same time, I confess to perplexity why something that seems very obvious to me has emerged as the point of contention.
Let me offer some other provocations.
My answers would be no, no, yes, no, yes. And in that gap lies my distinction between system and content. But now I am very curious as to everyone else’s answers.
With respect to grammars/machines I think I should just stress this point:
The traditional definition of a grammar or of a Turing machine contains the starting state and the possible transitions from every state to other states.
Thus, the starting positions of Chess are required parts of the grammar or machine of Chess, as grammars and machines are defined.
In using such machines and grammars as a starting point for defining games I think we can agree. I just think we disagree about what is a part of the machine or grammar.
In Chess, when iewed as a two-player game, no, the opponent is an external consideration to the definition of the game and does not affect it. The opponent chooses transitions in the game state just as you do. Viewed as a single-player game, however, a different opponent does make for a different game. Different chess AI’s may present very different challenges and as a single player game, the opponent will define the transitions between states (i.e. if you make move X, your opponent will make move Y). I would think that the two-player view is what we really mean by “Chess” and I would thus say that a new opponent doesn’t make it a new game.
I answered that before the kobold question because I think the same distinction applies. If you consider the game to be an entire system where stats are fluid then changing a kobold’s stats does not make for a new game. However, considered from a single-player perspective, yes, the transitions in the game have changed and it is a new game. The rules are fundamentally different. The rule used to be: whack this kobold 5 times with a newbie dagger and he dies. Now it is: whack this kobold 10 times with a newbie dagger and he dies. I would again say that I think the more global perspective is the one we tend to mean and as such I’d again agree that this isn’t a new game although I think that there is a meaningful sense in which it is a new game.
I think the polevaulting example is an obvious yes.
I think in the case of the Mario game, yes, from game theory standpoint this is a new game. The change is very subtle and I think it would be natural to recognize this linguistically by calling the two games by the same name even if they weren’t really the same. This isn’t Mario 1.09, instead it is Mario 1.10. There’s a distinct difference there even if it doesn’t merit a complete renaming.
Adding a piece to Chess obviously makes it a new game. I would say, however, that this is just as true if we move a piece to a new position on the starting board.
A question coming from the other direction:
Are all games using the Unreal Engine the same game? Even if they contain the same weapons and the same stats?
Raph wrote:
I would say: no, probably not (explained below), yes, no, no.
Changing the opponent in chess doesn’t change the game to not being chess. On the other hand, it can radically change the gameplay. I have to play at least a little different against someone with an aggressive opening game vs. someone with a strong middle game, for example.
I think that adding a piece to chess would still have the game be a variation of chess, so essentially the same game in the way that draw poker is the “same game” as stud poker. Not identical, but a majority of the game is the same. However, that change can have far-reaching implications. However, I don’t see this as being significantly different than playing against a different opponent.
Or, to draw from a poker example again, does requiring “Jacks or better” change the game? This essentially redefines what a “pair” is in terms of the card combinations, something that you could argue is a “mechanic”. However, this change doesn’t radically change the nature of game, in my opinion, and make it something completely different.
My further thoughts,
Perhaps it is time to introduce a notion of “variant” (which is after all a commonly used term) for games whose grammar shares substantial similarity with another. Then we can work on settling what exactly a “new game” means versus a “variant.”
Well, the Unreal Engine doesn’t impose that many constraints on gameplay, really…
Taken separately:
Two games both using the Unreal engine, no guarantees made as to any other commonality? Probably not the same game (though it is possible, if they have greater commonalities).
Two games both using the Unreal Engine, containing the same weapons and the same stats? I would say yes, if they also had the same freedoms of movement, victory conditions, and so on.
Two games of UT, using different maps? Same game, different content.
One game of UT playing CTF and the other playing Assault? Different variant games using mostly the same mechanics.
One boxed copy of UT: the Unreal Engine provided with a suite of mechanics and content that allows the play of several different games with the same materials.
What would your answers have been?
Naturally; that’s what I would expect content to do.
To me that’s a very clear case of a rule change, making it a “new game” though we can of course classify it as a variant.
As an analogy, is Tetris with hexes a new game? I think it is, because the rules have changed.
(Glad to see Hextris is still around, btw. š I played it first on the Mac way back when.)
I would say that game implementations with different stories and level designs but the same underlying physics engine are different games. The goal may be the same, “get to the last level”. The guns may be the same. However if everything else along the way is completely different and then the game states are completely different and the game itself is different.
Different guns as part of the example is probably a red herring anyway. Guns are content by your definition. We can change them in a spreadsheet to deal different damage, have different area effects, different rates of firing, different ammo capacities, etc. In fact there isn’t really that much different from passing a variable to a function which fires your gun and passing a function. In other words, rules can take other rules as parameters just as easily as they can take numbers. In a game like Magic:TG, rules that are parameters to other rules are common. I don’t think that lighting of a level is “content” and weapon types is a “mechanic” just because one is more easily expressed as a number than the other. Both have a very large impact on how the gameplay unfolds, how a player gets from one state to another, and as such they are both of the same class of mechanic.
Having to hit a kobold 5 times to kill it or 10 times to kill it IS a rule change.
Taking things to an extreme:
Most modern games are variants on C++. Programmers add “content” to C++ in the form of a text file containing code (which is encoded as a sort of table of numbers). But the only real mechanics of these games are C++ expressions.
And here, I think we bump into the need to define “game.”
I realized that often in this thread, I am referring to game as being ONLY the mechanics of the game, and sometimes, I am using it to mean the whole kit ‘n’ caboodle. That undoubtedly changes things.
So let’s back up (again)! Looking at some of the definitions of game out there, we can tease out some elements. Leaving aside all the questions of magic circles and the question of goals, and so on, I think we can agree on at least this part of the definition:
A game is a construct of rules.
I propose that for the sake of clarity, we refer to the bit that is just the rule construct as the ludeme (ludemes are built out of smaller ludemes, generally). Ludemes have no presentation; they exist in the abstract, and can be quantified or described by a description of the rules.
For those who have read Rules of Play, recall this bit:
I’ve been speaking of rules only, in this discussion, really. What’s more, I’m defining the ludeme (what I have been calling the mechanics) as what Salen and Zimmerman call “constituative rules,” which are
Then we have the sense in which changing the artwork on Quake makes it into a “new game” — which it certainly is in the layman’s sense. We can keep calling that “a game.”
A game is a ludeme with presentation. Two games might share the same ludeme top to bottom. And changing the ludeme but keeping the presentation the same is also a “new game.”
That would then satisfy your statement in this paragraph:
However, at that point, I would insist that a huge part of what goes into making those two implementations is really what I would call “experience design,” which would then be a facet of “game design.”
And we’d need a new name for the process of designing ludemes, which I will probably stubbornly slip and continue to call “game design” because I am completely persuaded that the ludeme is the part of games where the fun originates, and the part of games that is unique to games, and the part of games where true innovation comes from. š
I think that the form of the rules used to define a “game” can take many forms. However the one I would lean would be something like a context free grammar or a finite state machine (FSM). A very simple FSM is defined:
1) A set of states Q.
2) An element of Q designated as the start state q0
3) A subset of QxQ which defines possible state transitions among Q
4) A set of final states F which is a subset of Q.
FSM’s are frequently used to study games in Game Theory (the mathematical discipline). They can get a lot more complicated than that of course. In games of chance one might also ascribe probabilities of taking certain transitions. More complicated games may carry more state information around as the states are traversed (essentially becoming a Turing machine).
The important thing to me is that anything that changes the traversal among states is a change to the game as a theoretical construct. That includes such things as the starting positions in chess. The starting positions in chess are the base condition for a rather large recursion and are very, very important.
For example we can define factorial as:
for n an element of the natural numbers:
f(n) = n*f(n-1)
f(0) = 1
Without including the definition of f(0) we haven’t defined any function at all and if we define f(0) differently then we will have a function that is not factorial. If we don’t define starting positions for Chess then we have no game and we can’t study it formally. And, to my mind, it is clear that changing the starting positions is as much a change to the rules of the game as any other change we might make.
Similarly any content which changes the start of a game or the transitions from states is a rule change — a change to the formal system of the game. A game where it takes 10 hits to kill a kobold is different, at the formal level, from a game where it takes 5 hits to kill a kobold. The process of transitioning from the state of a live kobold to a dead kobold is fundamentally different.
Hmm… I definitely see your point.
However, what is f(n) = n*f(n-1) in isolation? The reason it is your example case is because it is recursive, and is predicated on a prior state.
In games, at each moment, the state you work with is given to you. It is essentially the output of a prior function (not necessarily the same function).
Just to toss a wrinkle into the chess problem example, the start state of chess problems can typically be expressed as a presumed chain of events that led to a given state, starting from the standard initial positions. The details of past events are irrelevant, save for those encoded in the current positions of remaining pieces.
In other words, given an initial data set, functions (we know not which out of the set of available ones) have been run repeatedly on the data, until we arrive at a given current state. Then we start the chess problem and proceed by applying functions — from the same set of available functions.
Unfortunately I’ve been out all day, so I’ve missed some good discussion here. But I’m going to revive this a bit and go ahead and tackle your query:
Now, let me first state that I pretty much agree with your distinction between ‘game’ and ‘ludeme’.
That said, after more thought, I agree with the distinction in the first two cases. Assuming that distinction, however, I still have to disagree with your last point in the above. Speaking purely in terms of ludemes, it is still functionally the same game.
Let’s say I wanted to give my young cousin an advantage and let him start with a few extra pawns. Certainly the strategy would change beyond words, but we’d still be playing ‘chess’. I argue that start state is not defined by ludemes but rather by ‘content’.
That said, you can’t have a game without content, so while the discussion on ludemes isn’t useless by far (and I agree, it also facinates me the most out of the four categories), it is also academic to consider it by itself.
That said, my answers to your query…
If you replace the word ‘game’ by ‘ludeme’: no, no, yes, no, no
If you leave the word ‘game’:yes, yes, yes, yes, yes
That said, this is the first time I’ve encountered the term ‘ludeme’ and I do like it. I’ve been meaning to use Levi-Strauss’ structural analysis of myth to do a structural analysis of the underlying structure of games, and I’ll have to borrow the term.
Another way to think of rules is as such:
The rules of the game should be all the things you need to know in order to recreate and play a model of the game. Graphical presentation and even story may not be necessary for this. But how to start and what form the content may take is. If I handed the rules of chess, minus the starting position, to someone who had never seen chess played, then they would have no real idea how to play the actual game of chess.
If you hand someone the rules for character creation and attacks but not the content of your world and ask them to recreate it you will obviously get something very different. I would say it’s obviously a different game although I can see how that might be argued. I think it’s still pretty much the same thing as the chess game. You may play something that shares certain rules as the original game but other rules will be far different and the differences will be extreme.
Excellent point.
So, I am willing to say that different starting positions for chess are different ludemes, most likely variant ludemes of the base premise. In the grammar, I consider the chess pieces to be abilities (e.g., representative of choices you can make), so that fits for me.
Hmm… and here, I am unsure. After all, this is exactly how pen and paper roleplaying games are played. You do in fact get something every different, but everyone still agrees it’s AD&D. What’s more, if you dug at it, you could probably reverse-engineer the right data ranges for the various statistics.
The statistics do not represent choices. They are things that you make choices about or against. So I don’t see them as resulting in a new ludeme. They certainly can result in a new “game,” particularly if paired with a new metaphor, as in all those D20 games.
Andrew Krausnick wrote:
I think this is insightful, and explains some of my answers more completely. A different opponent does change the game, but not the ludeme as Raph describes it.
Otherwise, great insight here.
Have fun,
The more you depend on the metaphor, the less likely you are to change the mechanics ā this is evident in the vast majority of titles developed today.
I noticed this quote on Darniaq’s blog. How can it make sense?
You say that the levels are completely independent. So you can take the same system and slap on it a different metaphor (the example of the healing game), as you can take the same metaphor and formalize it with different mechanics.
If the two revolve independently this means that they do not affect each other. So that quote cannot be correct.
Instead it’s when the two are strictly connected that they can affect each other and “coordinate”.
Here I imagined a combat system starting completely from the metaphoric level, but the result is all about the mechanics. How was it possible?
It makes sense for me because I still see them as dependent on each other. Yes, I could transform the fight in a puzzle game. It can be done, but instead I decide that the system should be *immersive*. Iconic, analogic. As similar as possible to what you could portray in your mind.
With no displacement between the metaphor and the meta-game.
And through the metaphoric level the mechanics WILL change. Probably so much that the idea can be not even technically doable.
What I mean is that the vast majority of games on the market today start with a metaphor or narrative, and copy the mechanics from elsewhere without really changing them one whit. This is the best way to minimize risk, because as risky as it is to create a metaphor and do a good presentation, it’s still a lot better understood that creating new mechanics.
I once played a Donald Duck platformer for the PS1 that was exactly the same mechanically as Crash Bandicoot. The jump timing was the same, the attacks were the same, and so on. It was clearly a different game, but the mechanics were identical. This is a very common thing, and I think we see it in the MMO arena as well.
This doesn’t mean that there isn’t some level of interdependence there, but it does mean that it sometimes gets ignored.
Hmm. Okay. Consider the Japanese Super Mario Bros. and Super Mario Bros. 2. For all intents and purposes SMB2 was “built on the SMB engine”. There’s a second player with variant physics. You can spawn a “powerup” that actually damages you. Going down a pipe doesn’t always take you forward in the level – this is a mechanic that showed up in World 8-4 of the original SMB but it’s spread outside the “final dungeon”. Finding a warp zone doesn’t always move you forward a world.
But there are no new enemies, no new capabilities, and the variant physics are basically “+ jump height, – ground friction”.
Are these different rules, or a variation on the existing rules? Can someone who is “played out” on SMB have fun with SMB2? Why or why not?
–GF
Hmmm….I think one of the mayor dificulties here is that within the framework Raph set up, certain old terminology is interchangable (ie can fit into more than one of the new catagories), depending on how they’re handled.
Lets take a game of UT2kX (not any earlier, I need the physics of the game intact to make my point :)).
Unlike StGabe, I think that if instead of ten hits it suddenly takes 5 hits to kill my opponent, I’m still playing the same game, only at a lower difficulty level. However, when we approach a limit, something happens: if it takes one hit and my target blows up in chunks, I’m playing an instagib variation of UT…and it is then arguable if I’m playing UT or a variation of UT or now a different game altogether! Personally I’d say I’m playing a variation of UT, as instagib in QuakeX and UT2kX are different things (due to other mechanics like movement and environment).
The same thing happens with gamelevels (content); take a base game of UT2kX and add one of the many usermade levels…you’re still undeniably playing UT2kX. Until of course you add some level which changes things to a certain degree, for example a level with huge unwalkable hills which you’d slide off…all other things being equal, you’d now be playing a game more akin to a strange cross between Tribes and marble madness. Content has become mechanics…
I wouldn’t be surpised at all if, in whatever structure, you would get some crossover/overlap; I’d imagine a piechart where the boundaries are fuzzy, and where one classification could cross over into the other if modified to a certain extent.
Glazius and MacD, I think the answer is the same in both cases. If the ludeme changes, it’s a new game. It might only be a variant game (certainly both your examples are just variants), but it’s a new game.
Examples of ludeme changes that were in the grammar (which I am going to link to again just to make sure everyone sees it):
I explicitly leave out stuff like “the shape of probability curves” because, as the presentation states, “mathematical balancing is a data issue, not a systemic issue.” I also state that “content is descriptive characteristics of a given problem,” analogous to adjectives.
So just a different map isn’t a new game.
I think we’re getting confused because we’re talking about a lot of different games with very different formal properties all at once.
For example:
See, now you are talking about a game that has difficulty levels which just means that we have added in some extra sorts of rules: if difficulty level == EASY then kobolds die in 5 hits else kobolds die in 10 hits. That’s a perfectly viable sort of rule to add to your game but it’s still a rule.
Now,
AD&D is a very different game than Chess. Whereas Chess is a very streamlined, your-turn-then-mine, win-loss-draw style game, AD&D is much more complicated. AD&D is a much higher-order process, it’s the sort of game that has rules that take rules as parameters (in a fashion somewhat similar to Nomic games). It’s transitions resist formal definition. But generally it is the same sort of thing. Different players, abiding by certain mechanics, provide statistical variance for the other players. Creating a certain dungeon for players is a much more complicated and less-restricted experience but, formally speaking, it has the same shape as creating a chess opening. And different DM’s, or DM’s over time, will provide very different, varying play experiences just as a different chess opponent or the same opponent over time will provide different playing experiences.
I think that, perhaps, the content-creation mechanics of games have become such an important part of some games that we want to grant them their own category of game-design. I think it makes sense to bring out their importance for some games but I’m not sure that they are indeed that different, formally, from other sorts of mechanics in more simple games. Also, I think it is a mistake to take from the importance of the content-creation mechanics in AD&D or MMORPG’s the feeling that content is somehow a core part of all games and not just a specific mechanic implemented in some games.
So. Grammar.
Mechanics = verbs.
Stats = adverbs.
Metaphors = nouns.
Narrative = adjectives.
Verbs can become adjectives (participles and gerunds).
Nouns can become adverbs (prepositions) and also adjectives (more prepositions).
And, of course, it’s in the modern tradition for a noun to get verbed.
–GF
[…] I’ve been thinking about a game based around amorphous and locked terrain. After glancing over Abalieno’s and Raph’s (among others) discussion of games, I believe the mechanic is amorphous terrain which can be locked by player action, and the metaphor is settlement of a wild forest by a vigorous and adventuresome people. (Mechanic and metaphor provide vocabulary, a scaffolding which allows me to explore the game.) Metaphor: the world outside is amorphous Not far beyond the golden fields of the village were serried ranks of forboding trees, the beginning of the wild forest. Good villagers didn’t venture very far into the darkness under the gnarled boughs, only much needed firewood and medicinal herbs could persuade one to enter that dim wood. It was rumored that trails shifted and that the way that explorers ventured out was rarely the way that they came back. […]
[…] over on raph’s site, he talked about making a healing game. a gentleman over on the cesspit has taken up the flag in argument against it. while his argument reflects some of the same feelings i have, i don’t think he’s conveyed it very well. […]
[…] A recent post on Raph Koster’s site , A bit on how I think games work, poses an interesting breakdown of game design into four layers; the mechanical, the statistical variation or content, the metaphor, and the narrative layer. […]
[…] Eric Zimmerman, Katie Salen (Rules of Play, Game Design Reader) https://www.raphkoster.com/2006/03/05/a-bit-on-how-i-think-games-work/ http://www.escapistmagazine.com/issue/41/9 http://www.alanemrich.com/SGI/Week_03/SGI_Week03.htm […]
[…] Koster argues that there are four stratifications in a game: mechanics, statistical variation, metaphor, and narrative. In Galatea, it is easy to see the […]
[…] a game grammar point of view, this is a clear example of getting the wrong end of the stick. Recall the distinctions between the “salad” and the “dressing” of a game — the […]
[…] Here’s your explanation, from my theory-of-fun/game-grammar point of view. […]
[…] is the core mechanic. That is what makes this medium be games, and not something else. (Source: Raph Koster’s Website) åäŗ«å°ļ¼ QQē©ŗé“ ę°ęµŖå¾®å å¼åæē½ […]