Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
July 28, 2014
arrowPress Releases
July 28, 2014
PR Newswire
View All

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

Life, loss and the beauty of the roguelike in Road Not Taken
Life, loss and the beauty of the roguelike in  Road Not Taken
October 18, 2013 | By Kris Ligman

October 18, 2013 | By Kris Ligman

Spry Fox of Triple Town fame has a new game on the way -- a richly-designed roguelike called Road Not Taken. Here, the independent studio's creative lead Dan Cook goes into detail about the inspiration behind the game, its influences, and creating an impressionist narrative players will remember.

You've drawn on poetic and art influences for Road Not Taken. Could you talk a bit about what drew you to these points of reference?

Dan Cook: As a general statement, language is not built to talk about systems. So we struggle to communicate aspects of games with metaphors and comparisons.

When I talk about my games, I find myself latching onto examples that are perhaps immediately evocative but require a little discussion to unpack. Pointillism for example is the act of building up a holistic image from a series of tiny dabs of paint. It came about due to emerging scientific theories on how light worked and how the eye perceived the world at a glance. We now think of these painters -- particularly earlier impressionists -- as profoundly emotional, but they came from a wonderful time where it was fine to use scientific thinking to experimentally push the boundaries of art.

Much of the discussion about stories in games isn't all that interesting. Instead I wanted to explore an alternate theory of how we generate stories from fragmentary experiences. Stories, to an even greater degree than sight, are a post processing operation. You experience something quite disconnected... sounds, sights, rushes of hormones, a pumping of blood. Often that sits unexplored, dribbling piecemeal into your long term memory as you sleep. Occasionally you tell someone about it or are prompted to remember it.

At this moment, a story is born. It is almost a wholly artificial entity and typically changes on retelling if not augmented by some technological feat like paper or video. The story leaves out 99 percent of what actually happened. It reconstructs critical details from sparse or missing moments. This is the nature of stories at a biological level, so what if we build games to fit that?

Games are experiences. They are a process players participate in. They are not pre-processed at the moment of being played. As such, maybe we don't need the hyper detail and coherency of a full blown novel to create a story in the player's head. This over specification was seen in the rigid pre-impressionist artwork of the age. What if instead we batter the players with micro-moments of emotion and visuals? A little snippet of text. An evocative moment between a mother and a child. Generate ten thousand of these moments using all our art, craft and algorithmic wit. Paint a cloud of experiential moments and then let the player craft what they will out of that.

This isn't particularly new... roguelikes and simulations have been doing similar things for ages. What is Dwarf Fortress if not a pointillist narrative?

In a way, the comparison of a game to pointillism is a very striking one, as you highlighted earlier. In a very literal sense all digital images work in the same way as pointillism does.

I love the term "pointillism" because it initially was an insult used by critics of the time, much like how armchair critics raised on the 3D worshiping propaganda of past console wars might sneer at the term "8-bit" or "retro." The rubrics used to judge obsessively-defined forms of the past are not necessarily appropriate to every new work. Nor for that matter should you ever listen to the current generation of critics. Wait for a couple crops to die off and cherry pick the ones who help pay your medical bills.

You mention in your recent blog post -- it was very evocative by the way -- how the game is about falling off the beaten path, so to speak. Gameplay-wise, it's a bit of a roguelike. How do you communicate a tone, or theme, if you will, if much of the player's experience will be procedurally generated?

I tend to think of the experiences of games as a probabilistic envelope of outcomes. Have you seen a pachinko machine? You drop a ball into it and the ball bounces around the pegs until it ends up in a slot at the bottom. Pachinko machines are obviously not a single linear sequence, but they are very much designed. You position the pegs, you say what lights up at what point. You run a vast number of simulations and test balls through the device and take note that none get stuck.

Each ball is a person. The collision of the ball with the peg is a person making a decision in your game. From the viewpoint of an individual ball, it has taken a completely linear path through a game. From the perspective of the designer, you are carefully crafting a probability space for how the game might be played. Too much tension? Turn that down. Not enough moments of loss and regret? Limit these themed resources a tad more and voila.

What is delightful about a roguelike or other procedural system is that instead of manually bending metal and repositioning pins in our machine, we can instead change a number and the whole contraption is instantly recreated from scratch. It is as if you had access to the fundamental physical constants of the universe and with a single keystroke you change the life paths of all future players.

Despite your blog post delving into some personal stuff, it seems clear from what we've seen of the game that it's fairly abstracted from your own life. I mean, unless you grew up around dire wolves and plague doctors. So I'm curious, what drove you to pursue a more mythic or fantastical setting as opposed to one which draws from the stuff of your day-to-day?

The things that go on day-to-day in my head are closer to dire wolves and plague doctors than autobiographical meditations on 90s punk bands. Also, you should visit northern Maine some January day.

On the practical side of things, there are a couple minor benefits to the current approach:

One, fantastical settings give a creator an opportunity to carefully peel back bandages on a sensitive topic. See [Minority Media's] Papa & Yo.

Two, they also offer certain escapist players an entrance into the game. I'm not completely convinced that a realistic children-freezing simulator has a large market. Perhaps that will be a future experiment. Do you think it would make a good Kickstarter?

How much are you involved in the game's art direction? It does sound like Road Not Taken is seeking to unify its visual and gameplay design a good deal.

The artist on the game is the extraordinarily talented Brent Kobayashi. To be honest, I let him do whatever he wants.

At Spry Fox, we build games with super small teams of very talented developers. It needs to be a collaboration and everyone on the team brings a big piece of themselves to the project. Our typical way of working is to write up some very short lists of what needs to be done and then trust everyone to do an amazing job.

The first pass can be rough, but then we critique it and see how it might gel better as a coherent whole. Usually after two to 10 passes, we've got something that works. It is a pretty ego-less process since folks share the ultimate goal of creating an amazing game.

So much of big team project management is about cutting or specifying. Yet, there's also a really important part of small projects where a developer is able to add little magical moments to the game. Because they were in that piece of code at the right time. And they saw an opportunity. And they had a rich enough understanding of the vision to add something that clicked in beautifully. And they had a personal inspiration that prompted them to create some glistening bauble of perfection. And of course they had the skill to pull it off.

With the wrong team, the most brilliant art direction in the world will fail to produce a wonderful game.

You can follow Dan Cook on Twitter at @danctheduck.

Related Jobs

Sony Online Entertainment, San Diego
Sony Online Entertainment, San Diego — San Diego, California, United States

UI Programmer, PlanetSide 2
Camy Electronics & Plastics
Camy Electronics & Plastics — Los Angeles, California, United States

Sales and Marketing Assistant
Red 5 Studios
Red 5 Studios — Laguna Hills, California, United States

Localization Project Manager
Red 5 Studios
Red 5 Studios — Laguna Hills, California, United States

Senior Gameplay Programmer


Keith Burgun
profile image
Very excited about RNT. Seems like it has a very interesting and original gameplay mechanism - which is true of all Spry Fox games, really.

One thing I wanted to counter-rant about is this propensity of - well, it seems like everyone - to want to mangle the words "narrative" and "story" until they have no utility. I've heard people try to argue that Chess and Tetris have a story before. If this is the case, then congrats: the word STORY now has zero use in the world of games, because literally every single game has it.

That should be a good rule of thumb for people. If you're using a word in such a way that it no longer excludes *anything*, then you're probably saying nothing.

(BTW: This might not apply at all to RNT, which may actually have elements of narrative and story in a way that people normally use the words. If so, then consider this a bit of a digression. Actually I should probably put this comment over on that Street Fighter story article, which also got me a bit riled up for the same reason.)

>What is Dwarf Fortress if not a pointillist narrative?

A fantasy world simulator. In fact, it probably *is* a fantasy world simulator, and not a pointillist narrative - or at least, it'll certainly fare much better when looked at as a fantasy world simulator.

Do you actually play DF? A lot of sessions are actually really boring. I've had long sessions go where nothing very interesting or dramatic happened at all - just a long series of somewhat rote tasks of digging, building rooms, etc. Of course, interesting, memorable things CAN happen, but you're not guaranteed anything.

Doesn't this sound more like a fantasy world simulator than it does a "narrative" application? A narrative DOES guarantee something interesting and dramatic. For a fantasy world simulator, it's actually to its CREDIT if some sessions are boring!

Daniel Cook
profile image
Ah, the dangers of a pithy summary statement that only has meaning in context of the preceding three paragraphs. :-)

- What am I saying about the inherent nature of story?
- What is the nature of the game as an experience and how does that yield story?
- How is the visual science behind pointillism a metaphor for this memory consolidation and social transfer process?
- How might all this tie into a deeper understanding of why Dwarf Fortress is a successful generator of artifacts such as Boatmurdered?
- More broadly how might you use this theory to create something people want to talk about after they've played it?

It is always pleasant to argue about semantics, however functional models of how the world works and tools that successfully yield desired outcomes are far more interesting.

Keith Burgun
profile image
You said that Dwarf Fortress is a pointillist narrative. If I push back on that, how is it fair to say that I'm arguing about semantics? I don't care about the words, I care about what you mean, and what I took your statements to mean is that you see DF as a system designed around narratives, rather than being a system that actually couldn't care less if it provided no narrative value at all. Yes: boatmurdered can happen. But many sessions are also calm, peaceful and uneventful.

We're talking about the PURPOSE of Dwarf Fortress here, and I'm saying that you're mischaracterizing its purpose by talking about narrative.

Further, I find this use of the words "narrative" and "story" a bit dishonest because anyone participating in ANY event technically "yields a story" in the same exact way. Why is it that when videogames "emerge into a list of events that happened", we call that a story, but we don't use that word for what the doctor did during your routine physical?

My answer is that we desperately want to associate the word "story" with games because most of us have serious doubts about the value of games on their own merits. We know that story is of value to people, and saying that games have story gives many people comfort, even if it doesn't mean very much because of how they've mangled the word to get there.

It's almost exactly like people who are clearly atheists who say they believe in god and by god they mean, "nature" or "something we can't explain" or something vague and unfalsifiable like that.

Kevin Maxon
profile image
I am someone who wants to say that Chess, and yes all games, create/tell/are stories.

This doesn't mean that the word story is meaningless. To show this, here are some examples of things that still aren't stories:
1) A single chess piece
2) The rules of chess
3) A single board-state in chess

'Story' used in this more holistic, systemic sense actually has much more utility for design, in my experience, than 'story' used in the traditional sense.

EDIT: Felt like I could write an entire article in response to this comment, so I did.

Kevin Fishburne
profile image
A story's what you make of the reasons for and relationships between unfolding events, regardless of what the events are. The events can be in your face, leaving little to the imagination, or more vague allowing you to interpret them more freely.

Your example of chess is also applicable to sports like baseball. Being a fan only for a few years, it took me a while to realize what the commentators were talking about when mentioning "story" while describing a ball game. While it sounds ridiculous to someone who doesn't know the game, it is true.

Mikail Yazbeck
profile image
"My answer is that we desperately want to associate the word "story" with games because most of us have serious doubts about the value of games on their own merits. We know that story is of value to people, and saying that games have story gives many people comfort, even if it doesn't mean very much because of how they've mangled the word to get there."

100% with you on that statement, beautifully put.

Kris Ligman
profile image

"We're talking about the PURPOSE of Dwarf Fortress here, and I'm saying that you're mischaracterizing its purpose by talking about narrative."

Dan might murder me for interjecting as a critic here, but -- speaking to what a game is INTENDED to do is hardly a productive line of argument. Dan's not saying that Tarn Adams necessarily *designed* DF as pointillist narrative, but that's the end result Dan sees it as having.

Arguing from a position of what a game is *for* is pretty shaky, honestly, whether you're in favor of seeing narrative as a broad concept or not.

Keith Nemitz
profile image
I want to call it "fragmented storytelling", but Dan's use of "pointillism" hits more true. It's all emergent narrative. I'll use the definition, emergent narrative occurs when gameplay takes over part of the storytelling.

The example I'm most familiar with is my 7 Grand Steps. The 'pointillist' stories within have little continuity, and no dramatic arc between them. Playing the game provides those elements of storytelling.

The difference between 7 Grand Steps and Road Not Taken is granularity. 7GS uses a grain size of flash fiction. RNT uses a grain size of sentences (or so I've been lead to believe).

My question is, would Bastion count as a 'pointillist' storytelling game?

Daniel Cook
profile image
For terminology, fragmented storytelling is a good phrase. 'Story bits' has come up as well. I often use the term 'evocative feedback' since it helps me distinguish between 'functional feedback' and hits upon how certain visual, textural, auditory, situational feedback act as a means of linking to or evoking a common experience. I tell you a story of hitting my thumb with a hammer and that stimuli triggers a cascade of associations from your own experiences that helps you empathize.

Bastion would absolutely count as an example. Short, contextual audio snippets are the fragments of choice there. I tend to use sentences and images. You use flash fiction. You could likely classify these evocative stimuli on a whole bunch of different axis.

- Length: Short or long?
- Priority: How much of the player's attention it consumes. Is it ambient or in the forefront.
- Channels: Layering sound, visuals, narration on top of one another.
- Generative: Is each bit authored or are variations generated using various algorithms.
- Frequency: How often does it repeat? Best practice is that purely evocative stimuli with no functional elements repeat once once fully comprehended in order to avoid burnout.
- Functional: Some of the best evocative feedback also has a functional element to it. Gun sounds are an example of one
- Quality: What is the perceived quality of the story bit.
- Uniqueness to the player: How much are the bits dependent on the unique actions of the player?
- Linearity / Dependency Structure: Can you play the fragments in any order? Or is there some form of dependency structure. Linear is the fallback. Bastion uses contextual events and reduces order dependency.

Each of these is a spectrum. I prefer to use a lot of ambient, short, templated images and text pieces tied into systems so they are also forms of functional feedback. I think of it as a middle ground that works well for the type of smaller projects I work on. The AAA game with cutscenes also exists within this model but shifted towards multi-channel, high priority, high quality, low uniqueness.

To continue with the pointillist metaphor, this gives us a painting kit. It doesn't tell us what to paint, but it does make it easy to understand the possibility space. Instead of replicating past patterns, we can assemble some of these pieces and try a new experiment with some rough understanding of how it might work out.

take care

Dene Carter
profile image
I think the whole story debate is slightly misguided.

A book or movie designed with a specific flow in mind is 'Story as Cause'.
*Everything* we do in life creates a 'story' of some sort: 'Story as Effect'. But it's usually crap.

Pure Story as Cause offers little replay value once all the narrative branches have been experienced.
Pure Story as Effect systems generate only very slight stories with little emotional engagement: a game of 'Go' for example.

Most entertainments offering narrative fall between the two, or occupy both, equally and simultaneously.

Story As Effect coupled with a rule system that is designed to generate meaningful events and a sense of narrative flow, can be argue to be 'proper' Story with a capital 'S'.

Before you react and say: "I've had some epic games of go! They had narrative flow!", here's a test: if something you're defining as a Story is not something someone would pay to read or watch or listen to, I'd argue it's not really a Story. A boring series of events that exhibit none of the skills of storytelling is not a Story, it is just 'events'. The word 'story' is so overused that I have mentally started thinking of it as 'story' with a small 's'.

So, yes, pretty much everything can be labelled a story:

"I ran out of cigarettes and will have to go to the shops."
"I am writing a comment that will cause someone to point out a logical fallacy."
"I'm thinking about lunch because I mis-spelled 'bugger' as 'burger'."

Yes, I've seen these kinds of things in movies and books...

...but they're not Story without a hell of a lot more added, and we're all smart enough to know that, really.