The content of this website is licensed under a Creative Commons CC BY-NC-SA 4.0 Attribution–NonCommercial– ShareAlike 4.0 International License. You can freely use and share everything non-commercially and under the same license as long as you attribute it to and link to:
Why DriveThruRPG? It’s the largest download store for role-playing stuff in existence and you’ll probably end up buying much more than just your copy of Ludotronics. Which would benefit game designers everywhere!
Why not Amazon? For one, illustrated non-fiction isn’t well-suited for the Kindle format. Also, at a €14.99 price point, Amazon’s cut amounts to €9.75. Well, no.
Level Five: Architectonics
Beat 2. Prime Positions
Beginnings & Endings
Whatever structure you end up using, you’ll always begin somewhere and end somewhere. In dramatic terms, you always need some kind of exposition and some kind of closure. Let’s look at both of them in the order in which they usually occur.
In Self-Editing for Fiction Writers, Rennie Brown and Dave King hammer home one particular point again and again: resist the urge to explain. Or, as they prominently call it, R.U.E. Related to “show, don’t tell,” on which there’ll be more to say in the following Beat 3. Paradigms Lost, R.U.E. highlights a problem that should be familiar to anyone who has ever played a dramatically complete video game. Namely, huge payloads of information clumsily packaged into dialogues early in the game to get the player up and running. Which, as everybody agrees, is a terrible habit.
There are four basic situations that pose different challenges, but also different opportunities for your exposition.
For the fourth and last case, some workarounds exist, but not many. There’s player character amnesia, as mentioned. Or a torrent of orders shouted at the player character by superiors. Or, it can be simply ignored, to the detriment of player motivation and player emotion. Most of the time, though, the game fails to resist the urge to explain and serves an opulent dish of unwholesome dialogue as a remedy.
Is there a recipe for the cold water situation that does not break the rule of R.U.E.? In a way, yes. You need to design a situation that transfers your exposition to one of the other three categories, be that a clever blank slate solution, a widening horizon setup, or some familiar or at least easily accessible ground that you create for the player within the unfamiliar world of your game. Should none of that work, there’s a dirty workaround you can use that breaks the R.U.E. rule in a reasonably acceptable manner: the roll-up. Or, in more familiar terms, the opening crawl. Your exposition doesn’t have to roll up or crawl, but it must throw the player within not more than three short paragraphs into the thick of the action. In the original Star Wars movie, everything flows from its opening crawl beautifully, and the two dialogues between Leia and Vader and Daine Jir and Vader reinforce and expand on the crawl’s expositional message. Imagine the information from the crawl’s three paragraphs had been crammed into these dialogues! What’s more, the audience wouldn’t have had a clue as to what’s going on for several scenes, except that a small spaceship and a big spaceship exchange fire with energy weapons and a contingent from the latter enters the former.
Perhaps you have realized already that this dirty workaround utilizes non-diegetic information that is not part of the game world. (Diegetic vs. non-diegetic information is discussed in-depth in Level Three: Plurimediality.) And it’s not just Star Wars! There’s a whole range of respectable movies that use this workaround, with Unforgiven’s two-paragraph roll-up as another famous example. And it works even better with games! Unlike movies, games already provide a lot of non-diegetic information through its interfaces throughout the game, so it stands out a lot less. Indeed, a succinct, powerful introduction that directly addresses the player can turn a cold water situation into a widening horizon situation. It can provide the player with enough exposition to create a familiar space in that world, and from that space you can lead the player into unfamiliar terrain at your leisure. Which, by the way, is even a highly desirable situation in general as it applies what Jesse Schell in The Art of Game Design calls J. R. R. Tolkien’s “distant mountains” technique. It creates a situation where there are always more things to explore and more wonders to see and more insights to gather just beyond the horizon of what the player already knows about the game world.
From there, let’s move on to endings. First off, games that have a match structure where the player either wins or loses against one or more opponents are not included. Chess does not have an “ending” in that sense. Nor, for that matter, have games like Skat, the game of Go, Unreal Tournament, match-based strategy games, brawlers, racing games, and so on.
If your game isn’t match-based, its ending will fall into one of four formal categories:
Perhaps you expected something along the lines of happy and sad endings, but these and similar classifications are too dependent on personal, social, and cultural meaning and context to be a good foundation for design decisions. How happily, from today’s perspective, does a pre-1960 movie end whose female lead character gives up her idée fixe (read: ambitions) and submits to becoming the male lead’s loving housewife ever after? Not very. From a formal perspective, though, this ending doubtlessly resolves the character development arc (and probably also the story development arc). Or, what if the character dies for a noble cause? Is that a happy ending? Not really. From a formal perspective, though, it resolves the story development arc, but not the character development arc (except that character displayed a conscious or subconscious death wish all along).
Here are the rules. If every development arc is resolved or every development arc remains unresolved, the game’s ending counts as resolved or unresolved, respectively. If at least one development arc is resolved and at least one development arc remains unresolved, the ending counts as mixed. If the game has more than one ending, depending on player performance including decisions (whereby each of these endings can appear resolved, unresolved, or mixed by itself), that counts as a multiple ending.
Let’s go through all four endings one by one, so you can decide which option is the right option for your game. (Right now, don’t worry about any of the upcoming parentheses with references to visible, internal, and proficiency goals; we will explore these and the nature of the game-driven goal in detail in the final Level Six: Integral Perspectives II.)
There’s nothing wrong with resolved endings at all. In dramatically complete games, it means that the world has been put back into balance or equilibrium (the visible goal from the story development arc); the player character achieves what they desire most (the internal goal from the character development arc); and the player has beaten the game (the proficiency goal from the player development arc).
Importantly, the character’s desire does not equal “inner peace” or some such, nor can it be superseded by it. When a character sacrifices their personal desire or even themselves for a worthwhile cause (like Rick in Casablanca or Jyn in Rogue One, respectively), the character development arc counts as unresolved (Rick still desires to be with Ilsa and Jyn still desires to live).
In games that are not dramatically complete, a game counts as resolved if it has a conclusive ending that is hard to achieve and the game stops for good after that. There are no more levels or enemies left, or tasks, and the player has achieved everything that can possibly be achieved. Think Desert Golfing after the 2017 patch.
In Swords & Circuitry, Neal and Jana Hallford argue that the player has put in too many hours into a game and worked too hard to deserve to get “hit” with a bad ending. Let’s think this through. To start with, an unresolved ending should never “hit” anyone. Any dramatic arc’s ending, resolved or unresolved, should make everything that has happened before worthwhile, relevant, and meaningful. For what it’s worth, this should allow the story and character development arcs to remain unresolved (the player character might even die, as in several famous and hugely successful examples). But what about the player development arc and player achievement? Here, Neal and Jana Hallford have a point. If a dramatically complete game is fully unresolved, i.e., the world’s balance isn’t restored, the player character’s desires thwarted, and the player has failed by design—well, that looks indefensible indeed. Even the dismal world of I Have No Mouth, and I Must Scream (which falls into the multiple ending category discussed below) offers the player a slim chance for a half-decent ending! Thus, fully unresolved endings are not recommended.
Games that are not dramatically complete are rarely purely unresolved by design. In arcade or F2P or management games or similar, it’s true that returning the game world to its equilibrium or balance becomes progressively harder until the player fails, so its development arcs appear unresolved. But there are high-scores and tokens and badges and similar design elements that provide a measure of personal achievement. When the player fails but scores a personal record, that failure doesn’t count as player failure. It can even be a towering achievement, like reaching the kill screen in Pac-Man or keeping a fortress alive for 200 years in Dwarf Fortress. Thus, as with dramatically complete games, endings for games that are not dramatically complete aren’t, or at least shouldn’t, be purely unresolved.
For a purely unresolved ending, and as a cautionary tale, take Desert Golfing before the 2017 patch. The game’s algorithms sooner or later created an “impossible hole.” That the event occurred was inevitable, but when it occurred was unpredictable. That way, the game world didn’t achieve equilibrium; the player failed by design, not through any fault of their own; and any achieved score was rendered meaningless since the impossible hole could hit the player at any state of achievement.
Like resolved endings, mixed endings are perfectly fine. In non-interactive media, there are two options. The world is put back into balance or equilibrium (story development arc) but the character fails to achieve their personal goals (character development arc). As an example, in the mini series The Pacific, the war is won, but the central character is broken (in contrast to its twin mini series Band of Brothers, where the war is won and the central character achieves his life goals). Or the other way around: the character achieves inner peace, but the world remains in shambles. This is rare in movies but frequent in television series; there, foregrounded characters in individual episodes often resolve a local character development arc, but the story world’s conflict remains unresolved (as it has to, in order to keep the series’s story engine running).
In games, interactivity and the player development arc make more permutations possible. Excluding forced failure in the player development arc for the reasons discussed above, there still remain three mixed ending options. The story development arc can remain unresolved; the character development arc can remain unresolved; or both story and character development arcs can remain unresolved. As long as there’s a resolved player development arc, all three options are on the table. (But remember, even and especially the most devastatingly unhappy ending must be relevant and meaningful for the player and consistent with everything that led up to it.)
Games that are not dramatically complete are, in a way, mixed by design; as discussed above, they almost never have purely unresolved endings thanks to high scores and other design elements.
A lot of ink has been spilled against multiple endings, and for good reasons. In The Art of Game Design, for example, Jesse Schell argues that not every ending of a story can be fully aligned, or unified, with the beginning of that story and its development over the course of the game. That is an excellent point. Then, others have argued that the player can never be sure to have gotten the best possible experience out of the game. That’s another excellent point. Finally, story forks are rarely what they’re cracked up to be as they almost always have a lot of game content in common. Which is understandable from the development perspective, but disappointing to players who try out a different fork in a replay. Also, publishers are terminally ill-disposed toward bankrolling content that only a few players might actually see, which is also understandable.
All these are very sound arguments against multiple endings, no doubt. Does that mean you should stay away from multiple endings?
Well no—but you have to do it differently. Let’s dig into this a little deeper. The trouble with multiple endings, as they are generally understood, is that they focus way too much on decisions that have different outcomes. If the player does A at some point in the game, event E happens, and the game proceeds from that event to ending Y. If, instead, the player does B at the same point in the game, the event F happens, and the game proceeds from that event to ending Z. Hence, this strategy forces you to twist your design into all kinds of contortions just to keep it from forking exponentially.
What you should do instead is focus not on what happens based on a certain player performance or decision, but how it happens. Roughly, and greatly abbreviated, if the player does A, event E happens, which leads to ending Z, but if the player does B, event E′ happens that has a slightly different quality and slightly different consequences than E, and the game proceeds to ending Z′, where things work out not as well as in Z courtesy of E′. This strategy, for example, is generously applied in The Witcher 3: Wild Hunt, where different players encounter the same assets in different states contingent upon earlier player decisions. Similar effects can be created by placing encounters and effects dynamically in cooperative games like Left4Dead, where an “AI Director” creates context-sensitive intensity curves. All this is not about “endings” but player narratives, and the resulting differences in these narratives spark curiosity and encourage replay.
Nevertheless, you can set your game up for multiple endings with different mixed ending combinations if you want to. In one ending, all three development arcs could be resolved. The world is saved and all your friends and team members are well. In another, one development arc could remain unresolved. The world is saved, but some of your friends die. Alternatively, all your friends are well, but the world can’t be saved. In yet another ending, two development arcs could remain unresolved. The world isn’t saved and some or most of your friends die. (But probably not all; you must leave some room for a successful player development arc.) That way, you have a robust set of multiple endings without the usual drawbacks. It’s a set that has a strong focus on player performance and player decisions; it can be developed with a bare minimum of forking; and it avoids all the major pitfalls enumerated above. Later in Beat 4. Problem Space, we will discuss a few more aspects of this technique in detail.
Neither ending is more right or more wrong than any other ending as long as it appears inevitable, in retrospect, with regard to everything that occurred before—game events, player actions, and player decisions. It can be a happy ending, it can be a sad ending, or everything in between. But it must make everything that has happened to the player during the course of the game worthwhile, relevant, and meaningful. What you want the player to experience at the end of your game is not an ending, after all, but closure.
Great beginnings and great endings are hard to pull off. Your exposition must lead the player into the game world without laborious explanations, which are detrimental to establishing emotions, but also without leaving them clueless, which is detrimental to establishing motivation. Your closure has to convey emotional truth and appear satisfying and inevitable with regard to everything that led up to it, especially player decisions. For both, you need a professional writer, which could be you. But still—later during the development phase, which is outside the scope of this treatise, it might be a good idea to involve an experienced screenwriter just for your beginning and your ending. Screenwriters are exceptionally well-trained in creating high-involvement expositions and closures that rock.
This phase’s reading order is fixed for the Introduction, Level One and Level Six.
Level Two through Five are non-linear: you can tackle them in any order you like.