Sunday, July 23, 2006

Ze Story Snobs

There exists a powerful group within the gaming world that actively seeks to stamp out innovation unless it falls along the prescribed lines of their rigid and conservative doctrine. Who are these people? They are every developer, gamer and publisher who promotes the ideal that a good game must have a story.

Raised on fine stories from the golden years of novel writing and movie production, story snobs see games as just another opportunity to tell great tales.

Poor David Jaffe fell victim to the bitter wrath of the snobs recently when he talked about how developing story based games isn’t all that exciting. It was an honest comment that makes a lot of sense if you’ve ever experienced the joy of tweaking a surprisingly interesting interactive system versus the slog of polishing a series of plot points.

It is really very simple. Not all games need stories. Treat story as one of many available marvelous ingredients that can improve your game, not as a necessity.

The logic of the Story Snobs
  1. I like games with stories!
  2. There aren’t as many great games with stories as there are books and movies with great stories.
  3. It is therefore the fault of [the developer, publisher, etc] because they are not filling my needs.
  4. As an advocate, I must passionately protect and promote any game with a story as the ideal.
  5. Anyone who suggests games without stories are reasonable should be crushed. After all, it is a zero sum game here. Any resources spent on promoting non-game stories are resources that could have been spent on 10 more dialog trees.
The root of this unfortunate attitude is a classic tale of old media infiltrating and co-opting a new media. There are three players:
  • The fanboys
  • The movie and book industry wannabes
  • The publishers

The Fanboys
There exist millions of fan boys who had great experiences with old adventure titles and Japanese-style RPGs such as Final Fantasy. These story rich titles were some of the first cross over genres that encouraged people not typically interested in games to pick them up and try them out. If you are a conservative media consumer used to movies, Final Fantasy is an easy dish to consume. You have to watch a few cut scenes, play a little bit of game and watch a few more cut scenes.

However, many of these new game players never moved onto new genres. Just like a good number of Brain Training new customers never try racing games, there are millions who started playing games with the adventure game genre and stopped when that market faltered. There are millions that to this day still play mostly Japanese-style RPGs.

For this demographic, the artificially sweetened formula of “Lots of plot with a dash of interactive bits” defines their total vision of gaming. As conservative media consumers, when game falls outside their nearly religious preferences they don’t merely accept and forgive. Instead, they are inclined to drag it behind their truck through the proverbial forums of Texas. “A game without a story? Impossible.”

Despite the copious evidence to the contrary.

The movie and book industry wannabes
I had a great conversation with an animator at GDC. He’s an industry veteran and works primarily on cut scenes. He confided to me that his true dream was to work on CG for movies. He read all the movie trade magazines and avidly sucked up their tips and techniques. He was in the game business because it was kind of similar and he could get a job there.

I’ve had roughly this same conversation with a remarkable number of developers. The game industry is filled with writers who want to author the next great novel, designers who want to direct the next great movie and artists who would be perfectly happy doing character design for a Saturday morning cartoon. Even if they aren’t actively trying to use the game industry as a stepping stone, many of their core values are informed by older, existing media such as movies or novels.

These cultural transfers from big established media industries have a huge impact on the type of games that are made. First, their general grasp of how interactive systems are built is quite weak. They couldn’t design a set of valid game mechanics if they tried. More importantly the passion for interactivity amongst many of the developers in the game industry is unexpectedly low. When you talk about making a sexy Blizzard-style rendered intro, eyes light up with respect and admiration. This is their dream. When you talk about emergent gameplay in a title like GTA, you’ll get blank stares. It just isn’t their passion.

If you have the skills to make movies, everything looks like a movie. There are a thousand decisions made during game development that are the creative choices of the developers involved. If your labor force is trained to build and steeped in the culture, and aesthetic of linear media, guess what most games will end up looking like? That’s right. Linear media with chunk of half assed or cloned interactivity thrown in for good measure.

I got a chance to read a game design document for a now published title. It read like a movie script. Except they had little production notes like “And now the character fights a red monster”. Interactivity in games should be more than just a production note.

But it never will be when large portions of our industry’s workforce worships the values of linear media over the unique charms of interactive gaming.

I can’t blame the business folks too much. They have their creative people telling them that stories are critical. They got violently passionate customers telling them that stories are the most important thing ever. So they do what sheeple do and green light mostly story-based projects.

The vast majority of the budgets in modern games goes towards art, video, dialog and other plot related expenses. The development teams are further stocked with Hollywood refuse, which only increases their story-centric biases. Game mechanics work is generally given less development time, resources or room for experimentation.

Since the production risk of story-based games is lower, publisher tend to green light them more often. The developers don’t know how to replicate the complex playground games that do become hits. The market ends up being flooded with dozens of story-based games and only a few games that focus on interactivity as the primary driver of value. So we train more players to expect story-based games and we train or import more developers that know how to only make story-based games.

The industry becomes more and more weighted towards producing games with stories. You end up with a feedback cycle that reinforces the required presence of story elements in most games. If everyone wants story, how can it be wrong?

As various folks have commented, Tetris would never be published today. The current requirement that most games must have stories is a filter that prevents the creation and publishing of what are potentially the crown jewels of the gaming industry.

There are lots of great games that don’t require a story. Focusing our effort on only creating games with story substantially limits our creative exploration of the media and limits the types of games that we, as game developers, are encouraged to create.

Stories are not required to make great games.
Before you think I’m a story hater, let me disabuse you of the notion. I like stories. I’m playing a darling little RPG right now called Aveyond that is quite plot heavy. Delightful stuff that simply would not work without the inclusion of a story.

Even as a story lover, however, the existence of the story fiends infiltrating every level of gaming irks me. They assume that stories are always a good thing. People are not thinking critically about whether or not their game needs story elements.

It is perfectly possible to have a great game whose plot elements fit on a postcard. Populous, Mario 64, Quake, Lumines, Bomberman, Guitar Hero, Counterstrike and hundreds of other titles succeed wildly as great gaming titles and yet all of them lack story beyond a rough setting. They don’t feed the player periodic plot points that extend a narrative. They don’t have characters with extensive histories that evolve and grow emotionally through a series of descriptive cut scenes. They don’t have fixed events that are described by a godly author as a way of informing the player about actions beyond the capabilities of the gaming system to simulate. And they rock none the less.

In fact, the one thing that prevents the game industry from turning into Hollywood with occasional button pushing to advance the plot is the fact that a lot of people purchase certain hits that shockingly have little evidence of tradition plot. Sports game and racing games consistently make a profit. Nintendogs and Brain Training came out of the blue and rocked Japan. Tetris made the Gameboy a success. All these smash hits have no plot and lots of interactivity.

So there is obviously a more complex tale to be told here. There exists a wide swath of games that can be successful without having a story. Just as there also exists a wide number of games that can benefit from having a story.

When players and developers simply assume that their games need stories because they have been blinded by their subconscious cultural biases, they fail to dig into the guts of why stories matter to games. When you say “Wouldn’t it be nice to have another cut scene because I like cut scenes,” you typically aren’t asking the hard question “What does this cut scene actually bring to the gaming experience as a whole?”

Story is a game design ingredient, not an end in and of itself.
Story has a purpose in game development. It is a ingredient. It has little inherent value by itself. Its primary value is how it contributes to the entire player experience. You are selling a game, not a movie or a novel. You need to design the whole game as a complete experience.

To use a bizarre analogy, making games is a lot like cooking. You may really like bleu cheese. I do. I once found a fabulous recipe for bleu cheese lasagna. The recipe called for a few crumbles of bleu cheese, but the store only sold large hunks. I thought to myself “I like bleu cheese a lot. Why waste all this cheese…I’ll just throw it all in!”


All the bleu cheese went in along with some expensive spices and other goodies. The result was a giant slab of goo that tasted intensely of bleu cheese. You couldn’t taste the spices, the sauce or the noodle. I ate it for two weeks straight and never ate bleu cheese again for months. I would have been better off just nibbling on the chunk of bleu cheese.

I added something in that I liked by itself, but I didn’t have a clue about how it would interact with or benefit the other elements in my dish. The same goes for gaming elements like story. What do they add to the game? If you end up with a game that is barely different from a movie, why not just make the movie in the first place?

What is a story to a game?
Let’s take a look at the role story plays in game development.

First off, it is worth defining story. Story is a series of linear narrative elements in the fashion of novels and movies from ages past. This is a very traditional definition that I’m confident does a disservice to many of the wacky interactive fiction attempts being concocted by mad geniuses around the world. It also happens to be the one that is most descriptive of the use of story in modern video games.

In most games story elements are used as rewards for player actions. The player does something and they get a little dose of plot. Typically plot points fall into one of three categories.
  • Enabling reward: These rewards help the player advance through the game further. Examples include the conversations in Half Life 4 that let you know that the main generator is down and needs a fadangle to fix it.

  • Red herring reward: These are rewards that the player instinctually pays attention to as potentially important, but in reality they are just tossed in there to help build a fantasy world. The player, not being able to distinguish between what clues are pertinent to the game world, laps the red herrings up and experiences the same sort of pleasure they would gain from an Enabling reward. Examples include descriptions of a Dark Past Foozle that once caused a huge cataclysm. It never affects the actual game, but players latch onto it and try to make sense of it none the less.

  • Visceral reward: These rewards trick our sensory system into thinking something interesting is happening. Example include big bloody fight scenes, spooky scenes that cause us to think we are in immediate danger even though we are actually sitting in a comfy chair in a posh apartment on the west side.

The model we are using here assumes that gamers are constantly trying to grok the gaming world in order to interact with it in a more meaningful manner. The primary bursts of pleasure come from activation of learning systems in the brain. There are secondary burst of sensation that come from false sensory input that activates various fight or flight mechanisms. It is a simple model, but it generally works and is a far better starting place than designing by feel alone.

When should story be used in a game?
So a story element is just a reward. It isn’t the only type of reward. It is one of many types of rewards. You could put in a cut scene when a big boss creature is destroyed, or you could let the player discover a new sword token that enables them to chop down the vines that have been blocking progress through the earlier jungle levels. Both might cost the same amount of development time and both are valid rewards that make the player feel great.

Instead of asking “how should I implement story in this game?” instead ask the question “What type of rewards best fit the game experience?”

Story-based rewards have several very distinctive characteristics that can influence your decision.
  • Triggers for specific types of emotions: The biggest benefit of story-based rewards is that you can use them to trigger social emotions such as sadness, humor or sympathy. These are typically difficult to trigger using algorithmic rewards, but are relatively easy to create using common narrative techniques and patterns.

  • Low initial production cost: With a line of text, you can hint at complex system that you never need to build. For example: “The tattered scroll describes the ancient history of Yendor where giant lavender airwhales ruled the skies” hints at a tantalizing other world that the game developer will never need to build. The cost? A few minutes of writing in a commonly available word processor. In general, a simple plot point can be created and polished at a much lower cost than what it takes to create an interactive reward.

  • Rapidly escalating costs as realism increases: As you attempt to increase visceral aspect of your story, production costs increase dramatically. Realism costs money in the form of expensive tools, talent and time.

  • Low execution risk: The risk of a story-based reward failing to be completed is very low. The production techniques for text, images, sounds and video are well understood and highly reproducible. If new technology is kept to a minimum, the use of story-based rewards is highly unlikely to delay the shipment of your game.

  • High burn out: Most story rewards have very low variability. When you see them once, you’ve sucked out 99% of their value. When you see them again, players get much less buzz. Repeated often enough and they become downright irritating. Imagine if you were forced to watch the main intro animation when you start up a game. It rapidly loses its appeal.

  • Limited economies of scale: With high burn out come very limited economies of scale. Every time you want a new reward, you need to custom craft a new one. The cost of generating new reward increases linearly with the number of rewards. More algorithmic systems, on the other hand, tend to have a higher initial cost but can be reused over and over again. Imagine having to come up with a unique and enticing story element every time the player killed a monster. It is much cheaper in the long run to simply give the player a few points or a health pack. Also note that the cost of a small increase in realism is multiplied across all the story rewards in your game. This gets expensive quicly.

  • Changes are expensive: Exploring variations on story elements is expensive. Often the plot points need to be rebuilt from scratch. If you are dealing with text, this isn’t so bad. If you are dealing with $50,000 cutscenes, it can be quite painful. Contrast this to more algorithmic system were changing drop percentages on rare items may be as simple as tweaking a single number and seeing what happens.

There are some folks out there who claim that stories work poorly with games. This has some root of truth. If you focus only on the quality of the story, you’ll find that your gameplay elements will appear to constantly interrupt and slow down the flow of the story compared to say your favorite movie. It can be difficult to built dramatic tension in a typical manner when the player is constantly jumping about, pressing buttons and performing other mechanical actions.

However, story can still be used as a vital element to the game. It can add an emotional richness to the reward system that is difficult and expensive to achieve using algorithmic techniques. Story must always serve the greater good of the game.

To return to our cooking metaphor, I find story to be much like a fine wine. When you pour a glass of gloriously rich merlot, you’ll discover all sorts of delicate nuances that are simply impossible to find anywhere else. Yet, that same glass of wine can also be used to cook some wonderful dishes. A nice lobster bisque wouldn’t be complete without a dash of wine to accent the flavor. Stories in games are really the equivalent of cooking wine, an essential and useful ingredient for many popular genres.

On the other hand, there are lots great dishes that you can cook without using any wine at all, just like there are some great games you can build that don’t use story. If game design is anything like cooking, there is an entire universe of game designs that work perfectly well without any story elements.

If you are yourself a story snob (I was for many years), you need to ask yourself “Am I in this business primarily to build a great game or am I in it to craft a great story?”

Pick your passion. If you really want to make movies, go for it. Move to LA, buy a video camera and get started! For those who choose to remain in the game industry, we’ve got a unique and wonderful medium that deserves to be explored and expanded as a powerful expressive force in its own right. Learn from old school linear media, but never be bound by its constraints. Use it as one ingredient in your dish only if you want a dash of story flavor.

Don’t be a story snob and assume blindly that your game needs a story. Players buy games for the total experience and you should choose the appropriate reward system that best fits the experience that you are attempting to craft.

Take care


How the story fiends filter great games
“I believe if Tetris were presented today, here is what the producer would be told: Go back give me more levels give me better graphics give me cinematics and you re probably going to need a movie license to sell that idea to the public. The producer would go away dejected. Today, Tetris might never be made.”
Satoru Iwata, GDC 2006 Keynote

Burn out: When you create story elements as rewards, you experience the same reward over and over again as you play test the title. This leads to burn out on those rewards.

“And the thing is, once you have the IDEA, your fun- as a designer- is really over. If you are working in the single player action-adventure genre, and are fortunate enough to be working with a team that can execute the crap out of what you think is an amazing idea, you don’t get much out of actually seeing your idea executed. You get a little, sure. You get the little tinglies and such. It’s a neat moment to see your idea brought to life. But you already saw the idea, already experienced the amazing moment...but it was in your head months ago. Now it’s just a slog to execute the damn thing so OTHERS- the PLAYERS- can enjoy what you’ve already finished enjoying.”

Lobster bisque with wine