In preparation for the NYU Game Center’s PRACTICE conference, we are publishing long-form interviews with some of the amazing designers that will be speaking at the event.
The PRACTICE: Game Design in Detail conference runs on November 9-11 2012. http://gamecenter.nyu.edu/practice/
In the fourth interview, MFA student Ilya Zarembsky interviews Daniel Cook, a veteran game designer who runs the popular game design website Lostgarden.com and is Chief Creative Officer at Spry Fox. His games include Triple Town, Steambirds, Tyrian and the upcoming Leap Day.
Ilya Zarembsky: In May, you posted on Lostgarden about an interesting side project you were working on — a web-based “3D modeling toy”. I would guess that the design of such custom tools is a vital and under-discussed part of a successful and efficient design process; could you tell us about a custom tool you used to help in the design of a Spry Fox game, and how using that tool helped?
Daniel Cook: The 3D modeling toy was more of a side project. I do art on the side and spent part of a side career making art tools too; the art tool design itch still comes up occasionally.
The truth of the matter is that we don’t use a huge number of custom content design tools, mostly because we don’t need them. My first rule of thumb is to eliminate content production in a design wherever possible. I’m a child of the great cost explosion where game development costs went from $10k to $100 million. In the process, we lost much of our creative freedom and became cogs in large uncaring corporate machines. So efficient content designs matter.
Often this means systems that provide lots of depth with very few assets (Triple Town). But you can squeeze in efficiencies by use combinatorics (like paper dolls) and other forms of procedural content generation. The actual content is often created use off the shelf hardware, off the shelf engines and a bit of code. Building complex tools to create simple assets can be another form of waste. Does your single piece of content scale to millions of uses? If so, perhaps it is worth creating.
IZ: Setting and narrative have been important in your games since Tyrian, a shmup (shoot’em up) that got lots of praise for having a surprisingly rich backstory. How do you go about finding and choosing a thematic wrapper for a particular set of core mechanics? How do you decide when a wrapper has become too ornate?
DC: There is a cost to the wrapper. And there is a benefit. Traditional media assets like text or images that convey most setting and narrative have a linear cost curve. The more you add, the more they cost.
The benefit of such content is twofold. The most basic is that it provides an entrance into the game. A setting triggers a particular mental schema that activates appropriate tools and skills. How do you impact the environment in this abstract game? Oh, it is a fantasy world and you have sword and that monster is charging you. You know what to do.
[EXPAND Click here to read the full interview.]
The longer-term benefit of setting is that it lets others tell stories about their experiences and share the lessons they’ve learned. Some call this behavior fan fiction or water cooler chatter. I think of it as a record of meaningful decisions and skills being transferred to the rest of the tribe. Games are abstract spaces at their core. Without a clear metaphor, the lessons of the rules and math are hard to communicate. Setting provides that metaphor.
What I’m constantly impressed by is how little you need in order to get these two benefits. So the question I’m always asking myself is “What is the absolutely minimum necessary to activate a desirable entrance schema? And what is the minimum necessary to give player tools to talk about the game with others?”
IZ: At first glance, a game like Triple Town seems radically different from old-school arcade games like Space Invaders or Pac Man. But I’d argue that Triple Town might actually have more in common with them than most recent arcade twitch games do. Unlike, say, Super Hexagon, it seems to have been designed with a very careful eye to what would maximize ROI given the intended play context. Would you agree? Could you talk a little about how knowing you were making a Facebook game influenced your design process?
DC: Triple Town was originally designed for the e-ink Kindles so I was very aware of the technical limitations. No animation, no color, size constraints, interface constraints, etc. So from the very start, it was a minimalist design.
However, there is a thing called the ‘content treadmill’ where you need to keep feeding a game content in order for players to keep coming back to it. In a game that runs as a service, having an expensive ongoing content cost can destroy a marginal business. And most indies are marginal businesses. 🙂 So for Triple Town, the content model built in exponentials so that a small amount of content could mathematically last for the lifetime of the universe. A mere 10 assets yields thousand of hours of gameplay by a highly skilled player.
IZ: Bunni is an odd duck in the Spry Fox catalog in that gameplay takes place in real time. Yet it feels like you can tab away from the game while the boss is visiting your department, come back to it fifteen minutes later, and feel safe that none of your efforts have been ruined. Can you share an insight or two about designing for variable-length, frequently interrupted play sessions?
DC: Spry Fox will have more real time games in the future. 🙂 I’m not really a twitch gamer so the intense style of real-time play doesn’t appeal to me much. For the last few years I’ve been really interested in games where you can pop in and pop out without too much trouble. Even Realm of the Mad God, which was very much an action game, included teleports so you could drop in for 30 seconds, jump to some players and get your fix. Then leave instantly.
I encourage designers to think about the logistics of play. When can people play? What is the chance of finding a concurrent match? What happens if someone has to go to the bathroom? What happens if someone has to leave the game for a day? Or a week? It is a rich vein for inventing new classes of game. We went through this exercise for our new game Leap Day and the result is a slow paced, real-time simulation game that has a rhythm quite different from most online games.
IZ: In one round of Triple Town, I survived for 55 turns, making 13 matches in the process – all the while using an RNG to determine my moves. To players weaned on games that punish your first mistake with death, a system lenient enough to allow this might seem like sacrilege, the antithesis of good hardcore game design. Yet at the same time, Triple Town clearly rewards skillful, deliberate play — when I played the level again, trying my best this time, I survived for much, much longer, and scored an order of magnitude more points. Could you share an insight or two on designing a system that can satisfyingly reward good decisions without frustratingly punishing bad ones?
DC: A certain class of concrete thinker often sees the world as black and white. I meet designers that passionately argue there is only winning and losing in games and everything else is a giant lie…for wimps. It is all a bit sadistic and status-seeking for me. I imagine young men trying to prove the viability of their sperm is the root of it all.
I tend to subscribe more to Raph Koster’s ideas that play results in the acquisition of skills. This is a much softer view. Practicing a game by playing ‘naively’ like you did in your example is a classic form of a play. You are watching how it works and getting a feel for the probabilities. When I was child, I would spend hours tying knots in shoelaces. I had no goals and I wasn’t even tying shoes. But the process was endlessly fascinating. Since certain complex patterns of cause and effect take many cycles to understand (especially when you are dealing with randomness or knot theory!), it is in the designer’s best interest to give you ways to explore those options cheaply and easily so you can slowly and eventually pick up on what is happening.
And then when you think you’ve got it, give overwhelming stimuli to let you know that you are on the right track.
Triple Town is a game that last a week and thousands of turns. In this context, destroying you after a single wrong move is counter-productive. It is like beating a child for playing with their dolls instead of getting a real job.
IZ: You often write about the importance of innovation, but you also champion a hard-nosed, pragmatic focus on profitability in the game design process. Your argument that focusing on innovation at the level of core mechanics is the soundest business strategy for small developers is very compelling, but I would argue that there remains a tension and a degree of incompatibility (at least in the short term) between the drive to make something truly new and the demands of the market. Have there been moments at Spry Fox when you’ve felt those forces pulling in different design directions? How did you negotiate that tension?
DC: The tension is always present. And I just ignore it since I know the cost of listening to it. For me personally, I will never be enough of a perfectionist to compete in established genres against large budgets. I don’t have the art, the design, or marketing skills to craft the best match-3 game. Nor do I have the patience. So when I go for the ‘safe’ route, I toss away my primary competitive advantage. Instead of having a slightly rough, but clearly innovative game, I have a crappy copy of something that already exists.
For example, as we were prototyping Leap Day, it obviously could have turned into a Tower Defense game. So we said “No…it isn’t going to be a tower defense game.” So we intentionally avoided making a B+ tower defense game in very crowded market. And in turn, we were forced to make something a lot more interesting and unique.
IZ: You’ve criticized the “I want to make a better X” game design approach for being insufficiently ambitious and flirting dangerously with plagiarism. Yet I would argue that many of your designs can be seen as “better X” games, or solutions to genre problems — the impersonality of shmups, the susceptibility of word search games to game-warping dictionary algorithms, the off-putting environmental mores of RTS games, the shallowness of the match-three genre. But to the degree that the games revise deep mechanics rather than superficial trappings and offer surprising insights and critiques of genre conventions, that doesn’t seem problematic. Any thoughts? When is a game so much better than its sources that it becomes something new?
DC: Players tend to see games in the narrative context of what has come before. So even if you invent the automobile, they’ll refer to it as a horseless carriage. In fact, the smart inventor will call it a ‘horseless carriage’ in their marketing materials because that is the theming that gives people a way to understand the invention.
Triple Town is a good example. It comes from prototypes that had very little to do with match-3 games. It has sets and matching but so do a lot of card games. Yet it doesn’t play anything like a match-3 game, which tend to be about a very different set of core skills. Instead of visual pattern matching and zoning out, Triple Town is about planning ahead, mastering randomness and managing space. If anything, it is a crafting / building game. However, by theming it like a match-3 and calling the ‘Civilization of match-3 games, voila we’ve got a doorway.
Our consciousness is a rationalization engine. Give it a game and it creates an entire backstory of why it exists and how it is related to every other game that it has ever encountered (like was done in this question). If the game eventually prompts someone to tell a story about its uniqueness then perhaps in time, it will be considered a new-to-the-world work.
Dan Cook is a speaker at PRACTICE this year. Join us on the weekend of November 9th – 11th for a deep dive into game design theory and practice with Dan and a collection of the best game designers in the field.