Communicating with Prototypes

For the last month, as I've detailed on this blog, I've been working on a project with the artist brontosaurus.

Both of us are very new at this whole indie game development thing, and we have made plenty of mistakes. And we will make plenty more. But we are learning.

I thought I'd share some of our latest mistakes, and what we've learned from them.

First of all, we've switched game ideas almost every week for the past several weeks. This in itself is not the problem. We change direction because we find problems with our old ideas that prompt us to start anew.

Here's a quick overview of the journey our ideas have taken:

We started with a game based on this picture.

Then we switched over to procedural aliens.

But we couldn't turn it into a game.
We tried to find a game here.

Then we decided to simplify.
We thought climbing would be easier.

But we couldn't make it fun.
So we switched over to suburbia.

And now, we're still in suburbia, but reconsidering.

Why have we had to switch ideas so frequently? What mistakes made it so difficult for us to just stick to one idea and finish it?

It has been a progression of mistakes. We started off trying to randomly doodle a game into a existence. Then we realized that we need to start with actions and gameplay before we start doodling interesting worlds and characters.

Now I'm beginning to see that our process for coming up with gameplay has been flawed.

My specialty is in game design, and programming. I choose to work with brontosaurus because his skills complement mine - his specialty is in world design, and art.

The way it usually works is that someone like me comes up with an idea for a game, and then finds an artist who is interested in making the art for that game idea. But we did it differently. Since I didn't have a particular idea in mind, and I like talking with brontosaurus about game design, we thought we'd just come up with an idea together.

This was not our mistake. The problem started because we were both communicating our ideas in words.

When I try to explain a game idea to brontosaurus, he has trouble understanding how it would look or feel. But the ideas I come up with usually have interesting game mechanics that I could prototype.

When brontosaurus explains a game idea to me, it's always very evocative and visually interesting, and he's ready to create all sorts of cool concept art for it. But I have no idea how to start prototyping it, since the mechanics are too vague.

Because I've been impatient to agree on an idea and get going, I always go with the ideas that brontosaurus comes up with, even though I don't know what the gameplay will look like. But sooner or later the vagueness of the design catches up to us, and we reluctantly decide to come up with a new, more feasible game idea. This is how the cycle continues.

Brainstorming may be fine in text. But when it comes to choosing ideas to work on, it's not really fair to evaluate our ideas until we have experienced them in either a visual or procedural form. I will express my ideas through prototypes, and brontosaurus will express his ideas through concept art.

If my prototypes inspire brontosaurus to come up with a world and an art style, then we can make them into full games. If brontosaurus' concept art inspires me to invent mechanics and gameplay, then we can turn those into games, too. But our starting point must be tangible. Words are not enough.

Let the game designers come up with gameplay, in the form of prototypes. Let the world designers come up with worlds, in the form of concept art. Don't force one to do the other's job.

It may take longer at first, but it's the only way we'll make something that we're both satisfied with.

Let's hope it will work in time for the contest. ;)

In other words, Less Talk More Rock.


Thoughts on the EXPLORE Contest

I'm working on a project with brontosaurus.

We have seven days to submit a game for the EXPLORE contest at Jay is Games. I like our current idea, and I think we should keep going with it. But even if we can put it together in seven days, I'd want to take at least another seven days to playtest and polish it. In other words, I don't want to rush this for the contest.

We could take a break and make a short, quick game for the contest just for fun, not expecting to win anything. I had a few ideas for that. One is to use our conversation history as the game content, since we have plenty of it, and it's interesting. There are a number of ways we could use it, but one very simple way to do it would be combine it with the Linear RPG. Or the typography could be a physical space to traverse, like Silent Conversation, with emoticons as powerups or something.

I was thinking about the Stick Figures story in Tales from Outer Suburbia, and where the story is, or where you might find a game there. It seems to just describe a scene, a place, without turning it into a particular story. It's hard to identify the protagonist, antagonist, and conflict. Maybe the antagonist is actually the confusion felt by the townspeople, rather than an actual character. Who knows? But you can more easily see little stories inside of it, like a kid smashing a stick figure with a baseball bat and getting frustrated, and uncertainty turning to fear and anger.

I was thinking about games, and how you could have a persistent world that is experienced through short, repeated game sessions. Little stories inside a world. Imagine Canabalt if instead of starting over when you die, you start again in the same world, at the place where you left off, or on the ground where you landed, or as another character somewhere else. It is similar to Calamity Annie, the way the narrative progresses even as you play again, where the gameplay repeats, but the world is fleshed out as you continually revisit it.

I'm also thinking about Zero Punctuation. Our conversation history could take the role of the monologue that narrates the images in a Zero Punctuation review. The only difference is that it is a dialogue, in text and not in speech. But the communication is similar. It does not tell a story, it discusses and explains. However, if we were to do this, we'd need a way to automatically come up with the images based on the text since it is not feasible to generate so much content by hand.

Freedom of movement across an image, linear movement along a text. Text forms the goals, while images are the medium of action? One action per line of text, one visual change. How do you create images from text? How do you combine text and action to get an image? The image and text give rise to the feedback. And the text leads the progression.

I was thinking about Spirited Away, my favorite animated movie. If Canabalt is like The Matrix, what would be the equivalent for Spirited Away? For Canabalt, you take the decisions that are implied and make them explicit. Or rather, you make it explicit, since there is only one decision. What equivalent can be found in Spirited Away? What decision can be made with the press of a single button?

But Canabalt is a linear game. You decide nothing. You act, in order to experience. Can exploration be about anything other than freedom of movement? When does exploration feel linear? A map does not dictate a path. When does it feel linear? A maze might have only one path, though it looks like a map. A labyrinth, even more so. Look at the feet. They never change their rhythm, yet the landscape around them changes. Is that exploration? What if they do change their rhythm?

You could make a game where you press a button to turn and change direction. This could be freedom of movement. Is it linear? If your actions have bigger consequences later on, maybe it is not really linear. If you turn here, you got to the desert. If you turn here, you go to the ocean. In Canabalt, your actions now have their consequence now - they do not change your route for the future. Can you explore without changing your route?

I've been thinking about tension before it solidifies into an emotion. I've been able to notice this tension already several times today, and let it fade away without latching onto anything external and feeding off the imbalance that results. I imagine there could be something similar for positive emotions. It all reminds me very much of Daniel Cook's Constructing Artificial Emotions, and I am intrigued by the idea of incorporating this into a game.

Where will this take us? I don't know. We'll see.


Active Sketch 02 - Ledges

Another sketch.

This time, a prototype. I made it to test out the movement controls for a climbing game.

In our effort to make a cool intriguing game, the artist brontosaurus and I have been coming up with concept art and procedural sketches that we could combine into something interesting. But we've realized that our approach is in need of a slight course correction.

Procedural trees and space invaders may be cool to look at, but that's not the same as a game that's fun to play. If you create a bunch of cool things, you don't automatically end up with an interesting interaction when you put them all together. But if you start by creating cool actions, things to do, then you're more likely to end up with something that's fun to play.

Daniel Cook put it this way, when asked for advice:

Typically what I'd suggest is working on a core mechanic and seeing if you can 'find the fun'. I see you focusing more on artwork...which is pretty, but doesn't find the fun.

Given this suggestion, I realized it might be more appropriate to spend less time creating procedural art and more time prototyping gameplay.

This doesn't mean that I shouldn't spend any time creating procedural art. It just means that before I come up with things, I need to come up with the actions first. If I have the gameplay down, the stuff you do, then I'll know what sort of things would go well with those actions, what would make them more fun or meaningful or interesting.

So this is our new doctrine: design the verbs first, then the nouns. Not the other way around.

Thus, this climbing prototype.

It's simple. Use the arrow keys to move the climber around. You can press up to jump, or down to drop from a ledge. The interesting bit is that you can hang on ledges. Maybe not quite as interesting as jumping off walls, but hey, it's an experiment. :p

I used the Flixel engine to make it. Flixel is awesome. It's perfect for little games like this, especially if you like pixel-y graphics. I wholeheartedly recommend it. I'll write a blog post about it sometime.

However, brontosaurus and I probably won't be taking this climbing concept any further. We had thought it might be a good idea to start with a simple action like climbing, and build up from there. And we started creating some concept art for such a game, which you can see here.

But after trying out some movement controls in this prototype we decided that the gameplay wasn't strong enough to carry an entire game. And that's the whole point of a prototype - to tell you whether the game is worth making or not.

So we're back to the drawing board.

And with some inspiration from Shaun Tan and The Little Prince, hopefully we'll come up with something even better. I'll let you know how it goes. :)


Game Idea Giveaway - Garden Chomp

...continued from The Game Idea Giveaway Thread

Request by freelanceflashgames:
  • what sort of game idea you're looking for
    Something fairly simple and easy to code, yet fun. The game would probably feature achievements and upgrades, but it isn't required.
  • what your goals are in making this game
    To make a good game that will be different from others and people will have fun playing.

The normal idea: Lava Land

The weird idea: Garden Chomp

In short, the idea is a simple arcade collection game where you try to balance your needs with those of the surrounding ecosystem. You are a little guy running around a garden, seen from above. There may be a few scattered rocks or plants as obstacles, but generally the playing area is unobstructed. Your goal is to survive as long as possible without getting eaten or starving to death.

You have an energy meter that constantly decreases. To increase your energy meter, you eat mushrooms. Just move around with the arrow keys or the mouse, and touch a mushroom to eat it. Like collecting gold in N, you have to eat these mushrooms often enough that your energy stays above zero. Otherwise, you die.

Where do mushrooms come from? To answer that question, we'll have to introduce you to your delightful companions in this little garden - the chompers. A chomper is basically a big mouth that can walk around - like Pac-Man with teeth and legs. Their entire function is to eat and produce manure. As luck would have it, chomper manure is the perfect fertilizer for mushrooms. As the chompers amble around the screen, they leave a trail of manure in their wake, which soon sprouts a delicious bouquet of mushrooms for your consumption.

To produce manure, chompers must eat. If they're especially hungry, they'll try and eat you. Don't let that happen - keep them well fed. Like an aphid, you produce honeydew, a sweet delicacy much enjoyed by the chompers. You can click the mouse or press the space bar to produce honeydew and toss it in the direction you are facing, preferably into a Chomper's perpetually open mouth. Producing honeydew subtracts from your energy meter though, so make sure you are eating plenty of mushrooms!

It's not that complicated. If you touch a mushroom, you eat it and increase your energy. If you touch a chomper, you get bitten and lose a large amount of energy. If you touch manure, you get kind of sticky and move slower for a while.

  • You eat mushrooms and make honeydew.
  • Chompers eat honeydew and make manure.
  • Manure gradually sprouts into mushrooms.
It's a nice little cycle. And the timing and conversion rates will take a lot of balancing in order to feel right.

Even though the game is about surviving as long as possible, it might be a good idea to break it into levels. The first level might have one chomper, lots of mushrooms, and few obstacles. The next level could have two chompers, and maybe the next level has two chompers and a bunch of obstacles arranged in a sort of maze. To beat a level, maybe you have to fill up your energy meter all the way, where each level might require a different amount of energy. Or you could just make it about surviving for a certain amount of time, though I could see that being annoying in a level-based game.

And there are plenty of things you could do with random goodies and powerups, changing how fast you move, making mushrooms grow faster, giving you temporary invincibility, or just giving you extra energy. Aiming the honeydew could be made easy or difficult. I'd probably go far something fairly easy, where the honeydew travels in a straight line until it hits an obstacle or goes off the screen. If a chomper is facing you with its mouth open then you can toss some honeydew straight in without worrying too much about aiming skill.

There could even be different varieties of chompers - some completely harmless, some that get hungry quickly and will chase after you, some that move fast, some that move slow. If you want, you even could do some weird things with Langton's Ant and patterns of manure. Maybe the player could push the manure around to change how the chompers walk. But I'd suggest starting with simple random movement and going from there.

Good luck! :)

Want an idea? Make a request on The Game Idea Giveaway Thread!


The Invader Ecosystem

Humans, invaders, and trees.

For my first project since going indie, I've been working with the artist brontosaurus on a little game. The specifics of this project change around a lot, since we're taking a loose, spontaneous sort of approach to developing it. Lately we've been heading in a space invaders-ish direction, and we've been trying to think of how we might extract some actual gameplay from the procedurally generated toys we've been playing with.

In order to focus in on something simple, we've decided to base the entire game off the feel of the concept art above, I Have You Now.

We thought that might mean creating a little ecosystem of invaders that the player, a human, could observe and mess around with. Like, the invaders eat each other, maybe reproduce a little bit, and maybe you can help some invaders fight off some other ones, maybe do a bit of genetic engineering, and it would be all cool and stuff.

But I'm realizing that it would not really be all cool, or stuff.


In general, the full artificial life approach with procedural terrain, trees, and invaders is both too big and too bland. It's too big for us to finish quickly. And it doesn't have enough human interest. We've got to put the feeling in it, the fantasy, the thing that makes it worthwhile to participate in the experience.

It's too big in a lot of ways, but the main problem is artificial intelligence - how the invaders behave. Adopt an Invader is an elegant design because the core behavior of all the invaders is the same - a simple flying, shooting ship. Additional behaviors and personality, like dropping goodies or flying in a certain pattern, are manually added by the players. Computers do what they're good at, and the players do what they're good at. Everyone's happy.

The problem with this approach is technology. We can't make Adopt an Invader right now, since we don't have time or resources to set up a database-driven web site. But we could incorporate the core idea into a single player game. That is, take generic invaders and manually add personality to them through playing the game.

It's like Pokemon. But instead of collecting the creatures so you can fight better, you collect them so you can start building onto them. Not physically, but mentally, in terms of behavior. Each invader is like a base for your LEGO pieces.

Now you have a goal - find and acquire invaders to build on, as well as the raw materials with which to build. And now we can start turning this into an addictive loop. And so now we're getting somewhere, with this game design.

We'll see where this idea takes us. Both brontosaurus and I like creating things, so hopefully we can turn the process of creating itself into a game. Creating, as well as collecting things with which to create, that is. It's a lot like what we're doing right now - collecting inspiration art, trying to build on that to create a game. We'll see. :)

Any thoughts?

walking along...


Game Idea Giveaway - Lava Land

...continued from The Game Idea Giveaway Thread

Request by freelanceflashgames:
  • what sort of game idea you're looking for
    Something fairly simple and easy to code, yet fun. The game would probably feature achievements and upgrades, but it isn't required.
  • what your goals are in making this game
    To make a good game that will be different from others and people will have fun playing.
  • what games you've made already
    whack-a-mole, a sort of falling game, an unfinished platformer, unfinished climbing sort of game, and pong
  • your favorite Flash games
    Onslaught 2, amorphous +, hedgehog launch, the last stand, bubble tanks 2, fancy pants, bowmaster prelude are just some that I could think of.
  • your abilities in game design, programming, art, and sound
    Good game design, decent programming, pretty good art, and alright at finding sounds.
  • your preferences in game design, programming, art, and sound
    I'd say most is art, then programming and game design. I can't make my own sounds, so I usually go looking for them.

The weird idea: Garden Chomp

The normal idea: Lava Land

In short, this idea is a simple arcade collection game (like Pac-Man) on a shifting field of lava. You are a little guy running around on the field of molten lava. Your goal is to collect the crystals that form as the lava cools, without getting burned by the rock-meltingly high temperatures underfoot.

It's basically an enhanced version of Acid Rain, a game I made several years ago for the TI-83 graphing calculator. It's what the game always wanted to be, freed from the limitations of the calculator's tiny screen and slow processor.

The focus of the game is on the environment, and how it changes. The entire map is a 2.5D isometric grid of tiles, and fits on one screen, without scrolling. Each tile has a height and a temperature. Cool temperatures mean that a tile is solid rock, safe to walk on, while hot temperatures mean molten lava.

Blobs of molten lava constantly rain down from the sky, as if spewed out by a nearby volcano. Each blob lands on a random tile, raising its height with the added material and raising its temperature to molten levels. Molten lava will quickly cool and harden to become rock, but you don't want to be standing on it while you wait for that to happen. Sometimes a blob will cool and turn into a crystal instead. When that happens, you can walk onto that tile to harvest the crystal.

The net effect of all these blobs raining down is to continually raise the height of the lava field, forming an interesting landscape of peaks and valleys. To compensate for this (we can't make things too easy) the whole field is also sinking at a constant rate. All tiles sink gradually, and when their height drops to the lowest level, they go permanently molten, submerged in lava. Of course, they can be raised up again if a new blob falls from the sky.

The fun of the game is in making your way through this shifting landscape, avoiding danger and plotting your path to the crystals that pop up. All you do is move, with the arrow keys or the mouse, and jump, with the space bar or the mouse button. To collect a crystal or a powerup, just move onto it. Your protective suit gets damaged when you touch a molten hot tile - the hotter it is, the more damage it does. If your suit gets damaged too much, you lose the game.

That's pretty much it. It's meant to be a simple game, like something you might find on Orisinal. There are a number of details you could add, like crystals of different types and rarity, shields that recharge, a time limit, powerups, upgrades, or achievements. But that's up to you. Attractive artwork and well-tweaked numbers will be important for this game, as well as making sure the controls are easy to use. I'd be glad to help advise on the design once you've gotten something going. Let me know. :)

Want an idea? Make a request on The Game Idea Giveaway Thread!


Active Sketch 01 - Invaders

An update! :D

As I mentioned in my last post, I'm going indie.

And for my first project, I'm working with the artist brontosaurus to make a game showcasing his unique blend of pixel art and intriguing world design. In order to make the most effective use of his talents, we are trying to make developing a game as much as possible like doodling random pictures.

This means that we're creating little sketches in code, implementing one concept at a time in order to mix and match them into, hopefully, a cool game. We talk about our ideas, brontosaurus creates some concept art, and then I take the art and try to turn it into something procedural.

We're calling these procedural things Active Sketches, and I've just released my first one.

This is just a tool to generate a bunch of random space invaders, based on the invader fractal. It's amazing what the human brain will do with bilateral symmetry. :p

It's fun though. You can find some nice pixelated images quickly with this thing. Need help creating sprites for your new graphing calculator game? Just set the dimensions to 8x8 and click Generate until you find some that you like.

Next up - fractal invaders, colored invaders, procedural trees...

Ahh, trees.


Going Indie

I've graduated, left my old job, and now guess what...

I'm going indie! :D

I'll be making my own Flash games full-time, finally.

I'll be using this as an opportunity to experiment and learn - about design, about programming, about collaboration, and about monetization. And I'll be focusing on rapid development cycles, to learn as much as I can in a short period of time.

As the great Daniel Cook once said,

Experience comes from finishing multiple times. By doing shorter projects, you get more finishes under your belt.

Starting projects is easy. Getting into production is easy. Spending other people's money is easy.

It is only when you go beyond that to wrapping up a project and trying to make money does reality hit and strong lessons are learned.

This is really an opportunity to put many of my ideas to the test - not only game ideas, but ideas about process and how to do things. I'm especially eager to test some ideas about using micropayments and metrics in online games.

To start with, I'm working on a project with brontosaurus, a talented pixel artist and map designer. We're trying out a new process for "doodling" a game, one piece at a time. Every day, brontosaurus will come up with a new concept, like a procedural tree, or a movement animation, and the next day I try to implement it in code. We're hoping that this will allow us to organically build a game out of pieces, and quickly try out ideas, in a similar way to how you might create something spontaneously awesome by doodling.

You can expect to see updates soon. :D

Wish me luck! ;)