Firewater Prototype 1

If you publish Flash games in some capacity and would like to fund further development based on this prototype, don't hesistate to contact me at


Mouse: Drag to move player
Mouse Button: Drop bomb
'f': fullscreen, 'escape': leave fullscreen
'm': mute/unmute sound (not in fullscreen)

GOAL: Douse everyone who is burning with water to put out their fire. The game won't acknowledge that you've done so, though.

Please wait for this to load - there's no progress bar. Click the flash app a few times to give it input focus.
There's no way to restart, so press F5 to refresh the page to play again.

Design Notes


   Fleshed out, this game would give players constrained resources for altering walls, their task being to douse all the folks on fire. The idea is underdeveloped. It's mostly just a graphics demo.

A Mini-Essay: Physics Simulations, Spectacle, and Precision, and of course, Civilization

1. Progressing Capabilities

   You couldn't have made a convincing water simulation on an 8-bit NES processor. Now we have hardware can handle it. Better still, more expressive langauages and better tool chains make it easier and easier. This fact is unambiguously progress.
   This increase of capabilities is mirrored in physics simulation in games generally. Many physics models were simply beyond us before. Now we can do more. All indicators point to a future where ever more simulation is in our reach. In fact, hardware companies are quick to stress this exact story.
   But a little devil lurks, quietly, in the background of this topic.

2. What Constraints Are We Breaking?

   Increasing capabilities is unambiguously progress. It does not follow, however, that because we can things previously unavailable to us, that we always ought to. David Hume and the is/ought split rears its head here..
   In a few scant decades, game development has overcome a plethora of constraints. To name just a few:

      2D Cameras -> 3D Cameras
      Clean Mathematical Player Control -> Animation Driven Player Control
      2D Playfields -> 3D Playfields
      Bitmaps -> Polygonal Models
      Stateless Games -> Pervasive World Games
      Disconnected Games -> Social Games
      Premise Driven Games -> Deeply Narrative Driven Games
      Digital Gamepads -> Analog Joysticks
      Turn based -> Real Time
      Grid Based Level Design -> Polygon Soup Level Design
      Simple Physics -> Complex Physics
      Game-y Objects -> Simulated Objects
      Game-y AI/Behavior -> "Realistic" AI
      Solitary Players -> Players with Chatty AI Buddies
      Quiet/Text Multiplayer -> Full Pervasive Real Time Chat Multiplayer
      Single Screen Games -> Scrolling Playfields
      Digitial/Discrete Button Input -> Motion Control...

   Every transition here has been heralded at some point as inevitable progress for game development. And yet, what would that even mean? How could we tell? Indie game development, currently ascendant, benefits hugely by being able to pick and choose which of this "progress" to adopt. The truth is, each pair above implies trade-offs, not progress.

3. The Role of Constraints in Other Media

   Technological advances in videos games, the increase of our capabilities, do have a one way arrow. This deeply confuses how we think about constraints, I think. Over in the land of words, the constraints of the sonnet were invented after people had mastered chatting in structured ways. The constraint was a liberating innovation. The formal bounds haiku were likewise generative. In music, the formal constraints of a canon allow for the exploration of unmapped musical territory. A 12 tone scale, or blues song structure, impose constraints, but those constraints are astonishing vistas of possibilities. Even in games, the evolution from kids kicking a ball around in a back lot to the formalized rules of soccer started, not ended, an explosion of complexity and detail and possibility. Other sports looks similar.

4. An Alternate History for Video Games

   We don't think that way with video game design most of the time. We tend to say, "In the beginning, in the bad old days, we couldn't do much... but then TECHNOLOGY MARCHED FORWARD, and now we can do things we couldn't do before! So we should!"
   Were the history of video game design reversed, and had we started with all these magnificent capabilities, I bet these constraints we've thrown off would be hailed as astonishing formal innovations. "Look at what happens when you fix a camera on a 2D plane with no rotation!" we'd say. "This is amazing - this constraint leads to such amazing kinds of games!" Maybe we've evolve there, once the novelty wears off.
   Until then, these consequences of our "progress" orientation are peculiar, especially for the following reason.
   We might upgrade hardware, and programming languages, and tool chains, and code libraries, and social ecosystems for capturing and sharing knowledge. But we can't upgrade our players.

5. We Can't Upgrade Players

   This fact brings consequences. People have hard and fast limitations on what they can understand about systems in motion, especially with real-time interactions. Putting 10000 enemies on a screen is a technical achievement compared to 10, but players probably can't make sense of this new space; it's all just noise. We might scoff at old grid-based game worlds and marvel at new, massively cluttered, hyper detailed environments, but players still must navigate those environments and make reasonable plans of action in those spaces. Enemies with pattern-driven behavior, like in Legend of Zelda: A Link to the Past might strike us as simplistic compared to the complex monster AI we can create now, but players still must understand the role of enemies in varying and evolving the play experience and building progression. Fully 3D cameras may be more cinematic and dramatic that the unmoving, single screen of Pac-Man, but players move their character in direction they expect in Pac-Man, and players in Pac-Man always see nearby ghosts when appropriate; overly enthusiastic 3D camera AI does not have the same track record re: player knowledge and intentionality. Motion controllers do open up new game styles and player expressiveness, but players still must feel that actions formulated in their head will happen in the game through their input interface. New procedurally driven player animation and control may be visually gorgeous, but players still need the actions formulated in their head to take place in the game world. And we may enthuse about ever-more-subtle physics simulation, and look down on older, simpler models of player, world, and monster collision and interactions, but players still need to have a reasonable model of how the game world acts and how their intentions and inputs will map into that system in motion. There is a tension in the heart of this all.

6. Abandoning Precision

   This tension has led to a specific evolution in games. The move is from precision to spectacle, from complex signal with minimal noise to simple signals and complex noise.
   Almost all the improvements listed above - complex physics, complicated 3D cameras, motion controls, hyper-detailed noisy environments, animation driven control systems, complicated AI and buddies - make games more complicated. This is especially true when all deployed at once. Had these features been added in early generations of games, those games would have just been wildly harder. But, developers introduced many of these shiny, highly sellable features exactly as large publishers doubled down on growing mass audiences. So, developers compensated by paring down the actual difficulty of these games. This move wasn't just to broaden appeal to less experienced audiences. It was inevitable to balance the innate difficulties of these newer features.

7. Dracula X and God of War

   Developers generated this ease by relenting on precision and de-emphasizing strict cause and effect in game designs. Perhaps examples will help. Dracula X:Rondo of Blood has a static, simple translating 2D camera. It has instant math-driven or extremely predictable simple animation driven input, highly discrete aiming and attacking, discrete monster movement and behavior, and grid-based predictable 2D environments. In Dracula X, accidentally succeeding at challenges rarely happens. Outcomes depend heavily on specific player intention and input.
   Compare this to God of War. God of War has a highly cinematic 3D camera, heavily animation driven input, very analog / imprecise aiming constantly reoriented by a rotating camera, and analog monster movement and positioning. The environment in God of War, with its rich artistic details, doesn't make nearly so clear where players can and cannot be, especially with an active camera. As God of War supports these complicated features, it is built to accept sloppiness in its combat system. Even good players miss attacks because the camera wanders, and because the analog joystick and player animation system together make player facing unpredictable. Players mis-time jumps for similar reasons. Players take damage in situations that don't feel like the player was at fault. To compensate, the game might have incorporated subtle player auto-aiming, for jumping or attacking - 3d actions games frequently do. Designers must've had this looseness and unpredictability in mind when designing the players attacks and abilities, and the attacks and abilities of enemies. They had to recalibrate and soften expectations about punishments for taking damage. Timing is adjusted (in an easier, more generous direction) compared to older 2D games. Interestingly, the art, animation, feedback, and narrative all strongly imply none of this. They say, God of War is a game of extreme brutality. But the mechanics aren't, and can't be, severe. The system doesn't support that quality. And so players absolutely can survive encounters in God of War without entirely understanding what happened. They can survive with button mashing. For some players, this is a feature, not a bug. Regardless, the relationship between player intention and outcomes is not nearly so tight in God of War as in Dracula X:Rondo of Blood.

8. Spectacle

   I call this the rise of spectacle in games. Positive feedback, chain reactions, and combos inundate players as they produce input only loosely connected to the specifics of the stimulus they're recieving. Those chains and combos are not events the player would or could have predicted, but they are flattering. This traits is rampant in newer games. And it's not just in games responding to these technical features. No one has ever, in the history of the world, accidentally cleared 4 lines in Tetris at once without knowing they were doing it , but it's rare game of Bejeweled that doesn't feature a chain reaction that the player didn't anticipate, especially with blocks dropping on screen mid-combo. It's the rare player who accidentally wins a level on The Incredible Machine without understanding that they're going to, but it happens constantly in Angry Birds, where arbitrarily hurling birds and seeing the result is core to the fun.
   Don't get me wrong. These new features do open up doors for the exploration of new types of games. And that's great! But I resist viewing them as blanket progress, rather than new tools in our toolkit.

9. Precision vs Accuracy

   In 11th grade chemistry, you might've been taught the distinction between accuracy and precision. In the context of measurement, accuracy is how close your measurements are to the actual value you're measuring. Precision is how close your measurements are to each other. You can have accurate measurements that are closer to the actual value than they are to each other. You can also have precise measurements that aren't close to a true value. The hallmark of accuracy is it's true to life. The hallmark of precision is it's predictable and repeatable.
   The transitions I list above are moves to "get closer to the real world". They improve accuracy. In real life, things are 3D, they mostly aren't on grids, orientation is analog rather than discrete, and movement is governed by complex physics and inertia rather than simple math equations. Developers keep pushing the limits of real-world accuracy in games further and further out. But the consequence is less predictability and repeatablility in games and thus, ultimately, less precision.