What Were the Odds?

Free Fall: Running with Armadillos


“If I have seen further [than certain other men] it is by standing upon the shoulders of giants.” – Sir Isaac Newton

I’m sweating profusely, concentrating intently and swearing under my breath. I’ve tried (and failed) nearly half a dozen times to move the orange armadillo from one end of the screen to another, and after nearly a half hour I still can’t walk away. I’m hooked. I have to get that damn armadillo home, and although the physics of the thing are as confounding to me as walking on water, every slight adjustment, every tweak of my machine brings me slightly closer to success, and I can see the path. I know I’m close.

The premise of Armadillo Run is simple: One must construct, using a small number of raw materials, a means of conveying an armadillo-shaped ball from one point to another. The challenge, however, is in building a structure than will not only withstand the forces exerted on it, but will also meet the rather strict budget requirements. Often, the solution is slightly more complex than it would at first appear, but that’s exactly what makes it so much fun – and frustrating.

I’ve constructed bridges, catapults, elevators and gravity slides, yet none of my contraptions are getting the job done under the budget allotted. And in spite of my curses, the damned armadillo refuses to break the laws of physics. There’s no cheating; either you build a device that will stand the test of the armadillo’s run, or you don’t. But even failure is fun, useful. It’s possible, through trial and error, to stumble on elegant solutions you would never have though of otherwise, and since it’s a game, nobody gets killed in the process. My only real problem with the game is why the hell anyone would choose an armadillo as a protagonist.

“The simple answer,” says Peter Stock, the designer of Armadillo Run, “is that the game design required the player to transport a ball and I wanted to have an animal theme to make it a bit more interesting. The only spherical animals I know of are armadillos and woodlice. I chose an armadillo because they seem a bit easier to relate to. Some girls I went to school with also had a thing for armadillos. Apparently they are the only mammals apart from humans to suffer from leprosy.”

To quote Bull Durham‘s Joe Riggins, baseball is “a simple game; you throw the ball, you catch the ball, you hit the ball.” Anybody can do it; few people are good at it. The same is true with game design, or, more specifically, designing a game as relatively simple as Armadillo Run. Everyone, on some level, knows the laws of physics, and everyone knows what’s fun and what isn’t. Few, however, can take the two ideas and make a great game out of them. Fewer still can do this alone, out of their house, and distribute the game, via shareware, over the internet. Peter Stock is among them. Seemingly the last of the “bedroom developers,” Stock took an idea, made it into a game, and sold thousands of copies. Just like that. Throwing, catching, hitting.

“My wife got a job that required her to relocate and I fancied a change,” Peter told The Escapist, “so I decided to try out making games instead of getting another ‘normal’ job. I had a few ideas about what sort of thing I wanted to do and I initially set aside six months to see what happened before reverting back to salaried employment if it didn’t work out.” A true Cinderella story, Peter had no previous game design experience, save working on a Tetris port. What he did have was an idea, a dream if you will, and a little computer programming experience.

“I studied computer science at university,” he says, “and I’ve worked as a programmer, but not in the games industry. I’ve worked on a variety of things, from fairly low-level search algorithms to 3-D modeling/rendering and database design/admin. Most of it taught me to have a clear, simple design, but the more immediately games-relevant parts of coding were the maths and techniques for optimization. Armadillo Run is my first proper game.”

Recommended Videos

Principia Mathematica
The screen is almost blank. The armadillo, his cheerfully blinking eye cast downward, rests frozen at the top of the screen. At the very bottom, hundreds of relative feet down, is the target. Newton’s Theory of Universal Gravitation suggests that the armadillo will rapidly descend, and without something to catch him, or break his fall, he will plummet off the bottom edge of the screen, and I will lose. Time to make a basket, then. A really big basket. I use metal rods and a big, flat sheet of rubber. Part of the structure implodes on impact, but it holds. I win. Thanks, Newton.

“The rules of Newtonian physics are pretty simple,” Peter told The Escapist, “yet they give rise to many complexities (and since most people are already familiar with them, there’s less to learn before playing the game).

“Many good games have this trait, but I think it’s tricky to design a set of simple, abstract rules that exhibit interesting complexity in gameplay. In this respect, the rules of physics are the gameplay rules, so I can’t really take the credit for designing them.”

What he can take credit for is using the laws to make a great game.

“I’ve had a few enquiries from some schools about it,” says Stock, “and I’ve received some positive feedback from them. I didn’t design it with [education] in mind, but it does seem to have some use as a teaching tool. I’m not sure if it’s used to teach physics or not, but I think Armadillo Run (like many other games) mostly teaches people problem solving and abstract reasoning.”

Scientific Method
Problem solving skills are exactly what I’ll need to develop to solve Level 22, “Croquet.” The level begins with the armadillo at the right edge of the screen, on one side of a long ramp. The target is at the other end. The ramp is level – no momentum. In the middle of the ramp, a few feet in the air, is a large anvil held stationary by the force exerted by a gigantic rocket. The rocket is on a timer; it will explode in a few seconds, and if the armadillo isn’t on the other side, it will be blocked, or worse, crushed. Time to make a mallet. And another. And another.

For a level with an obvious (self explanatory, really) solution, it’s easily one of the hardest in the game. Making a mallet with enough angular momentum to push the armadillo target-ward is easy. Doing so with the limited budget allotted is nearly impossible. My solution: five sheets of steel formed into a wedge push the armadillo down the ramp, just underneath the anvil as the rocket explodes. The armadillo is trapped at first, but as the anvil falls, it falls behind him, pushing him the rest of the way into the target. It took me about three dozen tries, methodically testing and recording and adjusting, but I made it. Thanks, Galileo.

“I [tried] to make the levels fairly open-ended,” says Peter Stock, “but knew it needed a purpose – a focus that would transform it from a toy to a game. Each level needs pass/fail/score criteria, and I think restricting the budget works well, since it allows a lot of freedom in the choice of design. I had some people comment that I should restrict the usage of each material, but I felt that would reduce player choices too much. From the solutions I’ve seen, I’m glad I didn’t do this – there are many interesting approaches I had never imagined that I would have inadvertently blocked had I done so.”

The materials in question are actually quite limited, and, of course, interesting. One can choose from metal sheets, bars, rubber, elastic, rope, cloth and (my favorite) rockets, when constructing an Armadillo ambulation apparatus, and each material has defined tolerances and costs, making a combination of all of the above the best solution. Need to build a straight, flat bridge between points? Some sort of metal structure supported by struts or suspended by ropes is the answer. Need to move the armadillo vertically up or down? Well, then, that’s when things get interesting.

“The rockets were included to spice things up a bit by introducing a source of energy,” says Peter. “Rockets are pretty fun to play around with, but I designed the levels so that rockets were only to be used on a few of them, otherwise there would be less incentive for players to use different approaches. I tried to make the pricing reflect their usefulness, so that different solutions were possible using a similar budget.

“Apart from the rockets, the materials represent each of the possible combinations of three properties (edge/center placement, flexibility and elasticity), so as with the armadillo, it was a case of finding suitable real-world materials to match those required by the game design.”

On the Revolutions of the Celestial Spheres
The starting screen is familiar: My little armadillo (I’ve named him Bob) is poised at the top edge of the screen, ready to plummet toward the target below. The only problem is that he and the target are not opposite each other; the target is about 20 feet to his right. When he falls, he will miss it – unless he’s nudged. Or slung. I construct a sling out of rope and cloth. The armadillo falls, he hits the sling, which rotates, spinning him out at just the right angle, just the right speed and into the waiting basket. Gentle as a kiss on the cheek. I’ve just constructed an homage to Copernicus.

Part of what makes Armadillo Run so much fun to play is the realistic way in which the materials act: metal sheets provide a rigid surface, add strength and weight; rubber bounces, of course, but also adds a subtle flexibility to structures; and cloth, as I’ve just observed, conforms to whatever it covers or contains, making it perfect for slings, pulley systems or to catch a falling object. I ask Peter Stock if he spent much time playing around with actual materials to get the feel of his in-game objects just-so.

“Well, I didn’t do any experiments with real armadillos!” he replies. “It was mostly a case of implementing some low-level physics laws and assigning realistic values to the various constants. As with the design of many games, I didn’t rigidly stick to reality in some places – sometimes for efficiency of implementation and sometimes to improve the gameplay.

“Although the behavior looks realistic, the gravity is too low – or more accurately, the scale of the objects is far too large (the armadillo is something like the equivalent of 1 m in diameter). I’ve read that this is a common technique used in games, to make them ‘more realistic'(!). This was done for both gameplay and implementation reasons – I think Hollywood makes us expect slow-motion action, and the physics calculations are easier to handle when objects are going slower.”

Like building a scale model of the solar system to spin at one’s leisure, just to see how the planets all move, by playing Armadillo Run we get a glimpse into the fun its creator had testing the boundaries of physics, game design and possibility. You can’t solve many of the levels in Armadillo Run by putting your armadillo on a sled and attaching as many rockets as will fit on the screen, but you can try that, if you want. You can try anything. But to advance, you have to tap the scientific principles and laws of physics handed down to us by generations of bright minds from the world over. Stand on the shoulders of giants, as it were. Just like Peter Stock did in making his game.

Sir Isaac Newton, who theorized that an invisible force (gravity) is what causes apples to fall down (and not up), was aware that those who came before him laid the foundation upon which his Theory of Universal Gravitation could stand. Among them was Johannes Kepler, who was the first to derive the laws of planetary motion (planets orbit the sun in a predictable pattern), yet did so by using observations made by the brilliant astronomer Tycho Brahe, who, in turn, like all astronomers and scientists since, built upon the observations and methods of Galileo Galilee who, for his own part, was building upon theories espoused by Copernicus. One inspired another, who inspired another, who inspired another. And so on, as the commercial says, and so on and so on.

Euclid came before them all. In the third century B.C., Euclid postulated that a single line could be drawn between any two points. Put a penny on your table. Then another one. Now draw a line between them. Simple. You’ve just proven Euclid’s first postulate, but you didn’t have to. He already did. Something so simple we now take it for granted was a radical line of thought in Ptolemaic Egypt, just as the fact that the earth was not flat was a radical departure 2,000 years later, and the idea that men could walk on the moon or that cancer could be cured have been departures in our own time. Theories which, we hope, will some day become so well known and accepted as to be taken for granted by future generations, who will quite possibly be traipsing between dimensions on their way to school. On Mars.

Euclid’s first postulate can (and has) been re-written as “the shortest distance between two points is a straight line,” making it the central tenant of not only geometry, but almost all modern thought. All game design, too. No one knows this better than Peter Stock, a man who stood on the shoulders of Euclid, Newton and everyone who followed in their footsteps to make one of the smartest, most addictive physics-based games ever created. The natural question then is, “What’s next for Peter Stock?”

“I had some ideas about making a totally different type of game, but since I have little experience in playing or making the types of games I was imagining, I’ve postponed those ideas for the moment, and I’m working on another physics-based game. The gameplay will be quite different from Armadillo Run – this time I’m going to incorporate some element of user control, and it will be more action-based, although there will still be the construction part of the game which encourages experimentation and refinement of design.

“I’m hoping that some improvements to the physics will enable objects to be simulated in greater detail, allowing players to create robots, which will be used to compete in certain challenges. If this is possible, then I’m imagining a ‘virtual Olympics’ type game with many different events.”

Robots … yes, we are enthusiastic about this idea. Very, very enthusiastic. From Euclid to Asimov, in two easy steps. From point A to point B.

Russ Pitts is an Associate Editor for The Escapist. He has written and produced for television, theatre and film, has been writing on the web since it was invented and claims to have played every console ever made.

About the author