The Power of N

Mare: We made time for it mostly after work, weekends…about 150 hours.

Raigan: That's like the good version. For the last three or four years we've been doing little experiments in physics and the collision detection, learning and screwing around.

M: That's true, we didn't just learn it from scratch.

Who did what?

M: Level designing were pretty much both of us, we do split up the coding but on this one Raigan did more of the coding and I did more of the art and interface stuff.

R: Most of the code was hooked into the physics stuff that I was screwing around with so it was easier.

So you didn't put comments on the code?

R: It started out so great! I had pages of comments explaining every process. But then I stopped. It got to the point where a bug would take a whole day to figure out, it was just such a mess. We totally learned that it's actually faster to take more time at the beginning.

M: Of course.

R: It feels great to rush through but then next week you're screwed. And it's dumb because we've read tons of post-mortems where they say just that…

Why'd you work in Flash?

M: It's so easy to use, it's instant gratification, you can throw in your graphics and instantly see how your code is running. It's fun and we already know it.

R: To write this game in C++ would take half a year. OK, it runs ten times slower, but if it means you can write it in a tenth of the time, maybe it's worth it in this age of super-fast computers.

M: Flash also allows you to compile for Mac and PC pretty easily and for free.

R: So many of the Flash games out there, they run at ten frames a second and they're not fun to play. It's very similar to programming for a 486, as long as you let that constraint shape your design. The design constraints of Flash - the renderer is so slow, and it's based on the area of the screen that changes from frame to frame. The reason why everything's so small is because it took up less area, and we could get more things moving at the same time: having five or six things small moving would be the same as two or three larger things are moving.

It is very minimal.

M: That was what we were going for. We were hoping that it would allow people to focus more on the gameplay. The character is pretty engaging, anyway.

The character is very expressive. Even when you die.

R: We made it fun to die.

M: We made it sort or random when you die, so it wouldn't get boring. Because you do die a lot.

I was really impressed with the way that when you blow apart, it's not just a death animation - bits of you bounce different ways, and you can see there's physics at work there.

R: Our way of doing physics is based on a great paper by this guy, Thomas Jakobsen (who did Hitman 2) -- it's a method that's really gaining popularity because it's driven by a simple, intuitive, computer-based approach to math, instead of the classical 18th century calculus-based approach that's a lot harder to understand.

What inspires you to put the time into implementing these constructs?

R: Physics allows for greater range of player expression. It's like comparing choose your own adventure to a sandbox. Instead of saying the player can to this, this or this, you make a system and they can do anything within the rules of that system. And that's another thing that's wrong with the commercial mode of production, that's why it takes so long - they have to produce so much content to predict what the player will do. They get interactivity by sheer volume of content. Physics isn't the only thing, it's a really well researched thing so it's pretty easy to implement. I'm excited for people to figure out how to implement new systems… it's really easy to program IF "X" THEN "Y" laws. But it doesn't take that much more effort to make it dynamic.

What do you have planned now?

R: We're going to release another fifty levels in another month.

M: Couple of months, maybe.

R: We're going to do a bunch of tutorials, release the source code for the collision detection…

M: We found a lot of stuff that helped us online so we kind of want to give back to that community. It's the same with releasing everything as free…

R: It goes against the model of how some people don't even write their games, they just decide which third-party licenses to buy and they plug them together and they have their game. From the perspective of the third-party developer it doesn't encourage them to share anything because then they'll be losing revenue.

Do you consider yourselves indie gamers?

M: I think so.

R: I think it was a huge thing for us when we met each other and learned about the Home of the Underdogs site. It made us realize, hey, we don't need to get jobs working for Electronic Arts, we can make games by ourselves that are really good.

M: Yeah.

R: If I was going to join a big programming team, I'd do it for applications - way less stress, way more money. I don't really understand why people work for EA. It's not their fault, it's such a huge engineering feat to make the games they're making that there's no way to do it but to use modern software engineering approaches where everything has to be by the book, it's an assembly line. When we were making N, stuff comes up - we just discover something that works - and we can decide right there, OK we're going to change the game. There's an artisan approach.

I know that when I'm writing a book, a character can develop in an unexpected way and I can take it in a different direction.

R: Whereas if you're working for a big company, they have to stick to the design document. Lots of things benefit from being made in that way, like making eyeglasses - it's great, there's an assembly line, it's the most efficient way. And there's some games, like sports games, you don't want them to be innovative - there are known rules for soccer - but right now, everything's being made under that model.

You can download N for free at Metanet Software.

Tags: assassins , collaboration , design , developers , freeware , interviews , Marc Ngui , Mare Sheppard , martial arts , ninjas , platformer , Raigan Burns , robots , simplicity , stealth