Nov 012005

Go read “How to Prototype a Game in Under 7 Days” at Gamasutra. Seriously. (I spotted it over at Game Girl Advance–thanks!).

To quote the article:

Handy Cut-Out List!

Setup: Rapid is a State of Mind

  • Embrace the Possibility of Failure – it
  • Encourages Creative Risk Taking
  • Enforce Short Development Cycles (More Time != More Quality)
  • Constrain Creativity to Make You Want it Even More
  • Gather a Kickass Team and an Objective Advisor – Mindset is as Important as Talent
  • Develop in Parallel for Maximum Splatter

Design: Creativity and the Myth of Brainstorming

  • Formal Brainstorming Has a 0% Success Rate
  • Gather Concept Art and Music to Create an Emotional Target
  • Simulate in Your Head – Pre-Prototype the Prototype

Development: Nobody Knows How You Made it, and Nobody Cares

  • Build the Toy First
  • If You Can Get Away With it, Fake it
    Cut Your Losses and “Learn When to Shoot Your Baby in the Crib”
  • Heavy Theming Will Not Salvage Bad Design (or “You Can’t Polish a Turd”)
  • But Overall Aesthetic Matters! Apply a Healthy Spread of Art, Sound, and Music
  • Nobody Cares About Your Great Engineering

General Gameplay: Sensual Lessons in Juicy Fun

  • Complexity is Not Necessary for Fun
  • Create a Sense of Ownership to Keep ’em Crawling Back for More
  • “Experimental” Does Not Mean “Complex”
  • Build Toward a Well Defined Goal
  • Make it Juicy!

A lot of these elements were things that became very apparent during the writing of AToF. As I wrote the book, I also spent a lot of time designing puzzle games and board games. The successful ones had some elements in common:

  • I limited my interface.
  • I had a strong central theme (“Make a game that looks and plays like a kaleidoscope feels.”)
  • I could prototype them myself in three hours.
  • I could prototype them on a piece of paper. I’ll write below about my game prototype kit.
  • I could polish the game over a weekend.
  • I always started with a “blue squares” demo of just the core mechanic–no graphics.
  • I spent my art time on feedback, not just dressing. When pieces are captured off a board on the screen, it’s nice to make them pop with a sound effect. It’s even nicer to have a ton of particles. Best yet is to have a gun slide in from off-screen, shoot the pieces into messy bouncing bits, and play a loud satisfying explosion sound.

To my mind, the practices described in the article are exactly how you learn the basics of game design. And losing sight of them is exactly how you lose sight of fun.

Raph’s game design prototype kit

  • Two decks of regular cards.
  • One deck of Uno cards.
  • One Go board.
  • One Checkers board.
  • A half dozen six-sided dice.
  • One full set of polyhedral dice.
  • A large stack of differently colored index cards.
  • Twelve pounds of differently colored beads. Go to the pottery aisle at your local craft store–these are the kind that get put in fish tanks and potted plants. It’s a bit more than a buck for a pound of one color.
  • Wooden pieces, also from the craft store. These are found in the aisle with the clock faces:
    • wood cubes, various sizes
    • colored flat squares, three sizes
    • dowel rods
    • ‘pawn’ pieces
    • wooden chip (circles)
    • assorted circles, hexagons, stars, etc
  • Blank wooden clock faces that you can draw boards on.
  • Wood glue
  • Dremel tool
  • Square glass chips (also from the craft store, asst colors)

I keep it all in a chest I bought for the purpose.

On the PC: I use BlitzBasic. Either Blitz3d or BlitzMax, usually BlitzMax. Lua is also good for the purpose, though it’s syntax drives me nuts.

  11 Responses to “How to Prototype a Game in Under 7 Days”

  1. So this is an inspiring article. And Raph Koster adds some commentary. What I’m worried about now is publishers saying, “Hey look, these guys can prototype a game in seven days. How come you need six months to do preproduction?” But not really. One interesting thing in the article is the anti-brainstorming stuff.

  2. will be selected for a paid summer internship with the company. Cheers to Kyle for the party invite and to the other Kyle for keeping me laughing throughout the presentation! You guys rock! =) For more details about the GDC presentation, check outRaph Koster’s blog where he expounds on the concept of rapid prototyping and even offers up a “game design prototype kit” to use when fleshing out that killer game design idea. Cheers, Raph!

  3. Blogroll Joel on SoftwareRaph Koster Sunny Walker Thoughts for Now Sex, Lies and Advertising

  4. of list of random pointers they picked up during the development. It’s more concentrated on how to create a prototype with a team and how to host the event. I know this isn’t directly related but for the sake of completeness here’s a link to Raph Koster’s blog, where he gives an interesting alternative approach to gameplay prototyping. There was also an interesting article about Paper Prototyping on There was a lot of talk about prototyping in this years

  5. Bartle?s 5 most important folks in virtual worlds [IMG] Posted by Raph’s Website [HTML][XML][PERM] on Fri, 20 Jul 2007 03:12:38 +0000

  6. Very interesting, Raph. I tend to be a bit more minimalist. I just cut squares out of paper for tokens, or use coins. Glass beads are also nice.

    One comment about programming, have you tried out Python instead of Lua? It’s fairly good for rapid development, and the syntax is really nice, especially for people that aren’t hard-core programmers. (The hard-core programmers tend to dislike the syntax, even, but I’ve found people take to it easily.) You can find information about the game development side of it at if anyone’s interested.

    Have fun,

  7. I used Python back in the Privateer Online days and for a while thereafter. I thought it was a really cool language, but I never found a way to throw up an animating sprite on screen in five lines, which is what I really wanted. Maybe PyGame can do that now, I don’t know?

  8. I’ll admit that I’ve never used PyGame, but from what I’ve read it seems to fit the bill.

    I really like Python because I was able to write a moderately complex program the same afternoon I started reading the language reference. The tools are nice, and the language syntax is very clean. Admittedly, I’m a programmer by training, but it seems the language was really easy to use.

    My further thoughts,

  9. Indeed, you can do that kind of stuff with PyGame these days. The initial setup is still a little bit of a pain, but once you get everything plonked down, you can whip out some pretty impressive prototypes in very little time. And like Brian says, you don’t need to be a rocket scientist to use it – I use Python and PyGame for the “Intro to Game Programming” class that I teach at ACC, which is geared towards people who have no desire to become programmers whatsoever, and I think it works out just fine. (Now that I think of it, here’s the resource page for the class – it should save anyone who’s interested a bit of time collecting all the bits and pieces…)

  10. Yeah, just the fact that you need a page like that is why I ended up at Blitz. It cost $80-100 (depending on which one), it was cross-platform, came with a very straightforward IDE but you could obtain fancier ones, it exposed OpenGL, and putting a sprite on screen involved:

    graphics 800, 600
    ball = LoadImage("ball.png")
    drawImage(ball, 300, 300)

    Nice and high-level, because it was designed specifically for game-making.

  11. […] I don’t know if he has the IP licensed, but it’s apparently Hobby Lobby Online. Posted by j at 08:42 PM […]

  12. […] We don’t need to just invent mathematically ideal formal systems, but we need formal systems that are also appropriate for the simulation we want to render. The myths and expectations are stronger and have an higher priority than the abstraction and perfection of a formal system. The idea of a “twitch” combat will remove the weight of the previous HUD-heavy gameplay that used a very high level of abstraction that was nowhere direct and intuitive (I wrote something similar while “reviewing” what we know about Vanguard). What is sure is that, if the implementation is good, the players will surely reward this choice. Apparently it was difficult creating special content for 34 different classes, which I’m guessing anyone can see is a ridiculous balancing and content creation task that never should have come about in the first place. “We looked at the professions and said we have 34 widely disparate professions in the game. None of which get all of the attention that a profession deserves in a game this type and many of which don’t resonate to the Star Wars universe. I mean, what is a pikeman and why is it something in the game?” So with that thought in mind, they’ve folded all of the 34 professions down into nine professions including Jedi, bounty hunters, spies, officers, smugglers, commandos, entertainers, traders, and medics. “We ended up with these nine very iconic character types. […]

Sorry, the comment form is closed at this time.