Canabalt (2009)

Endless runners use procedural generation a lot, because it’s hard to make an endless game without it. The game that popularized endless runners is no exception.

Canabalt carefully calibrates the buildings it generates ahead of your runner so that every jump is possible, even if it isn’t always easy. Where some games have their generation tweaked for maximum challenge, it’s obvious that Canabalt instead optimizes for maximum flow, the mental state where you lose yourself in the task you are performing.

There’s an updated version available now on Steam and mobile, but the original one-button web game is still available. 

Procedural “paintings” for the novice programmer.

(Another submission: and a really cool one. Starts from zero and programs an app that paints.)

fyprocessing:

(via Maze Bounce on Vimeo) from literacola

Experiments using maze generation as a sound visualizer.

Unusual inputs, everyone! Just because the label says something, doesn’t mean that’s what you need to use it for. Here’s sound being fed into a maze generator.

Using an input that already has an order to it can give a procedurally generated work a more engaging structure, with patterns layered on top of patterns. What if you used a maze generator as input into your sound synthesizer? 

Risk of Rain (2013)

A co-op platforming shooter with roguelike flair, Risk of Rain uses procedural generation to enhance it’s challenging luck-pushing mechanics, adding uncertainty and risk to keep the punishing gameplay engaging.

The objective on each level is to find the teleporter, activate it, and stay alive long enough to kill the remaining enemies. This is all the more difficult because the difficulty of the game goes up over time, leaving to a trade off between finding enough powerups and leaving the level before it’s too difficult to survive.

While certain overall aspects of each level are fixed, a significant number of the subfeatures are randomly shuffled, plus the location of things like powerup chests and the teleporter. 

This keeps it play much more fluid than a fixed level would be: there’s enough uncertainty that your play choices are about both the larger system and taking advantage of the details of this particular instance. Rather than optimizing a best run-through on fixed geometry, the players must switch up tactics to match the situation. The parts of the level that are fixed provide a structure that experienced players can use to anticipate possibilities while never having a certainty. 

In a game that’s all about the risk/reward balance, that’s a perfect use for procedural generation.

Waiting for Bárðarbunga (2015)

Art installations can use procedural generation too: generative art has quite a bit of history. Here’s an example: inspired by the possible eruption of the Icelandic volcano Bárðarbunga, François Quévillon created this video project that uses footage of the site and the equipment monitoring the site. The display recombines them, using the unusual input of the computer that is running the installation. The fans and power draw of the machine are translated into the non-linear presentation of the video production.

http://www.francois-quevillon.com/html/en/node/332

http://www.creativeapplications.net/openframeworks/waiting-for-bardarbunga-non-linear-landscape-cinema/

Bird Species Generator

I’m sure you’ve encountered this problem before: you have a pressing need for a new species of birds, but you’re not quite sure what it should be. Well I have good news for you! Anke has created a Bird Species Generator.

The names are convincing, in part because they are generated via recombination of existing bird species names. In an accidentally Borgean touch, there’s no check for existing bird species. There’s the possibility of generating a real bird species, embedded in a birding guide from the universe next door. Double check before you claim that the saffron-margined oasis waterthrush is entirely fictional.

http://lab.ankewehner.de/random-generators/bird-species/

Thanks to the person who (anonymously?) submitted a really cool video of generative cell growth to the blog. If you’d like to suggest something interesting you’ve made or found, check out the submission form.

Simulate the growth of cells and you quickly get tissues.  Next step is to combine tissues into an organism.

The Swallows

If you want your computer to write a novel, why not program it to simulate something like a text adventure?

I wouldn’t call it great literature, but of all the novels that came out of NaNoGenMo 2013, I think the most influential was Chris Pressey’s four-book series of The Swallows, The Swallows of Summer, Swallows and Sorrows, and Dial S for Swallows. It immediately generated a comprehensible plot, complete with a MacGuffin, and its structure suggested a number of ways the idea could be built upon–and I’ll be discussing those projects in the future.

Do peek at the Python source code if you’re interested in how it was constructed.

https://github.com/catseye/The-Swallows

⋆✵tiny star fields✵⋆

Bots don’t need to be limited to words or pictures: here’s a bot that makes happy little star fields for your Twitter timeline, entirely out of unicode characters. It’s a good reminder that the form the output takes is only limited by your imagination. You can sometimes get a surprisingly effective result from something simple but unexpected.

@tiny_star_field is from katierosepipkin, half of the duo behind Moth Generator.