You can't beat learning from personal experience... but learning from someone else's experience is often less painful. In this post, I've collated some of my favourite post mortems, game design documents, and design analyses for mainstream games, from Mario, Sonic, and Zelda to MGS2, The Sims Social, and Portal.
The NES was the Wild West of game development, I thought, lawless and free. [...] As it turns out, I was totally wrong! Instead of finding something outdated with a ton of nostalgia value, I found an excellent primer in the fundamentals of non-linear game design.
Tony Garcia and Mike Stout play through the Ratchet and Clank games they worked on, discussing the creation of the games as they do.
There are hours of these videos to watch, so here are a few moments that stuck out for me:
- On focus testing: They don't dumb down elements because focus testers can't figure out how to use them, they cut them out when they realise they can't afford to put in the time and resources to get the resources right.
- On side-quests and mini-games: "If the player didn't sign up for the thing you're designing, you really shouldn't make that thing super hard."
- On focus testing levels in "block form": It's really hard to get testers to look past unfinished art and test the gameplay alone; it invites simple criticism of "it just looks unfinished".
- On puzzle design: It's more important for a puzzle to make the player feel smart than it for the puzzle to require the player to be smart. (Also touched on in this video.)
- On hard vs. fun: "It's really easy to make something hard, but making something that's fun and difficult is different."
- On pathfinding: Tony discusses getting the Tyhrranoids to behave.
- On jobs: Mike and Tony discuss the state of the games industry, and getting a job in it.
By allowing the players to travel freely through time and space, the developers opened up the game world to exploration. Although most optional narrative sections are inaccessible until the player finds the Epoch - a time machine which also allows for fast travel through the game world - the player is allowed to find their own way through the main narrative with minimal interference.
The player's mechanical actions - a slow climb, gaining a small bit of ground every time she hits her jump key and manages to grab the next power-up token - mirror the protagonist, jill's situation. and central to the experience is upwards motion (reinforcing the metaphor, each token is an arrow pointing up). jill is trying to climb out of the underground and into the light; the player is jumping her avatar higher and higher to unlock the level exit.
The lesson for future games might be this: make your technology extremely simple, easy to modify, ship it with a diverse enough pool of content that people can extend it to create a variety of settings and styles, and promote the sharing of this content as a way to add value to your game.
I know this quote is about Doom, but I can't help but think of Minecraft.
Whenever a Metroid player aquires a new power-up, her mind races back in time in a way not unlike what happens at a turning point in a movie. When a secret is revealed we are forced back through the story to mentally review everything we've seen so far, sometimes changing the interpretation of entire scenes. So that's why Obi-Wan was so worried about Luke facing Vader. What did this change? This happens in Metroid too.
I don't agree entirely with Campster's analysis of Sonic games' mechanics, but I did find this a really interesting video nonetheless, with some excellent points.
Creating truly engaging software experiences is far more complex than one might assume, even in the simplest of computer games. Here is some of the cognitive science behind why Angry Birds is a truly winning user experience.
My informal opinion was that Angry Birds is an incredible collision of game design concepts, and it works. It's got a pinballish sort of recklessness and a Super Mario level of depth. It's a game - like Peggle - in which you make quick decisions by way of a user-friendly one-touch tool rife with nuance. In manipulating Angry Birds' slingshot, the player will, in less than one second, choose just one option from a pool of literally hundreds: how hard are you going to shoot that bird, and at what so very precise angle upward or downward? Players will not hesitate to make these decisions. They'll stretch that slingshot and let it snap, again and again. I'm the guy who stares at a restaurant menu until my date calls her mom or a taxi; I'll sit and stare at my new Skyrim characters for literally an hour before I can dare to start the game. Decisions paralyze; Angry Birds doesn't. I fling those little birds all over.
This is ostensibly an Angry Birds review, but really it's a ZiGGURAT post mortem.
Super Mario Bros. 3 contains many obvious design lessons that are also present in other games, e.g., the gradual layering of complexity that allows players to master a specific mechanic. What surprised me during my playthrough, though, was how some of these lessons were completely optional.
For example, it's possible to send a turtle shell skittering in the opposite direction of destructible bricks, or to take the cloud-route and skip certain powerups and interactive objects. Of course these same lessons are repeated multiple times, but they're not always as heavily hinted.
An hour and a half of Portal and Portal 2 writer Erik Wolpaw discussing his work.
Chris DeLeon was frustrated by the number of Super Breakout clones that got the game "offensively wrong", so he made his own. This blog post details the differences between the original version (and his own remake) and the many other versions of Breakout, explaining why each design decision is so important.
In this brief post, Edmund McMillen explains how he and Tommy Refenes took the traditional "difficult platformer" formula and refined it for Super Meat Boy.
The feel of steering Mario around in Super Mario 64 fills me, to this day, with thoughtless joy. Especially in Bomb-Omb Battlefield where there's very little pressure or structure, I love to just run and bounce and spin, experiencing the sheer kinetic joy of controlling Mario. Control, intent, and instructions flow from me into the game as quickly as I can think.
Feedback returns just as quickly, letting me adjust and fine-tune my instructions. When a game feels like this I'm hooked, ready to spend endless hours discovering every nook and cranny. Considering the near-universal reverence for Mario 64 among both players and creators of video games - to say nothing of the millions of copies it sold - I think it's safe to say I'm not alone in enjoying the feel of Mario 64. As a developer, though, I have to wonder: wherein lies the magic? What's behind the curtain? A huge part of it is the feel.
I'm pleased to note that the MGS2 GDD is as pretentious as the game itself. (And I say that as a massive fan of the series.)
Note: This is for the original GTA, working title "Race'n'Chase", not any of the modern 3D blockbusters. If you enjoy this taste of history, consider buying Jacked: The Outlaw Story of Grand Theft Auto, which tells the story of the series from Race'n'Chase to GTA IV.
There are more GDDs and similar documents at Game Pitches - I've just picked my three favourites here.
A deep article about the cynicism of modern "social" games that have themselves been "gamified". Also see the companion review of The Sims Social at actionbutton dot net.
Got Any More?
This is my personal collection of favourite game design analyses, post mortems, and GDDs, bookmarked over several years. I know there are many more out there that I haven't come across yet - in particular, I'd like to see more about indie games - so if you know of any, please share them in the comments!