Generative Landscape

World-making and terraforming

21. Daniel Brown generates dystopian housing projects in his beautifully lit fractal series, Travelling by Numbers (2016).

Brief

Write a program that presents an ever-changing, imaginative “landscape.” Populate your landscape with features that are suitable for your concept: trees, buildings, vehicles, animals, people, food items, body parts, hairs, seaweed, space junk, zombies, etc.

Give consideration to the depth of variation in your landscape: after how much time does your landscape become predictable? How might you forestall this as long as possible? How can you generate a landscape that is both coherent and engaging?

Consider: foreground, middle-ground, and background “layers”; variation at the macro-scale, meso-scale, and micro-scale; natural and human-made features; utopia, dystopia, and heterotopia; the immersive use of motion parallax; and the potential for surprise through the placement of infrequent features.

Learning Objectives

• Apply principles of generative design to terrain, scenery, and worlds of the imagination
• Bias randomness to carefully regulate probabilities
• Carry out a metadesign process

Variations

• Populate your landscape with one or more of the “three verticals” (people, trees, and buildings): according to Jungian psychology, these are the defining psychological features of landscapes.
• Pay attention to the manner in which the landscape moves past the “camera.” For example, it might appear to scroll by (as if you were looking out the window of a train); or approach from a first-person point of view (as if you were driving, or riding a roller coaster), or slide underneath (as if you were looking out of a glass-bottomed airplane). Consider a moving or even roving camera, capable of rotation as well as translation.
• Depict an outside scene, an interior one (such as objects on a conveyor belt), or an altogether dreamlike one.
• Experiment with 3D (as in noise terrains); 2D (as in side-scrolling video games); “2.5D” layered spaces; orthographic views; or even nonlinear, non-Cartesian geometries.
• Give consideration to sound and the possibility for audiovisual synchronicities (as in Guitar Hero).
• Make an autonomous creature, vehicle, or other character traverse your landscape.
• Implement features in your landscape that grow, evolve, or erode over time.

Making It Meaningful

We are a migrant species, instilled with a wanderlust that continually clamors for new horizons. Before the modern era of mobility, landscape paintings were often the primary means by which people could visualize faraway lands and mentally escape to them.

Today, eight-year-olds trade “seeds” for favored Minecraft worlds, and procedurally generated environments have become commonplace in video games, where the algorithmic production of novel landscapes is an economic necessity for inexhaustible play. For the meta-designer and artist-programmer, there is assuredly something godlike about calling forth world upon world. It is probably not a coincidence that the first all-CGI sequence in a feature film depicted the synthesis of an entire planet, in the triumphant “Genesis Sequence” of Star Trek II (1982).

Generative design systems, whether used to create faces, landscapes, creatures, or chairs, define seemingly infinite possibility spaces. Pay heed, however, to what Kate Compton calls the “10,000 Bowls of Oatmeal Problem”: “I can easily generate 10,000 bowls of plain oatmeal, with each oat being in a different position and different orientation, and mathematically speaking they will all be completely unique. But the user will likely just see a lot of oatmeal.” ⁱ As Compton indicates, the challenge and opportunity of meta-design is in architecting systems whose results offer perceptual uniqueness, and are thus meaningfully distinct.

This assignment asks you to bring forth a world from your imagination. Alternatively, you may create an accurate computational representation of a very real place—and generate “more” of it.

22. Kristyn Janae Solie's Lonely Planets (2013) is a stylized 3D terrain that shifts between minimalism and psychedelia. The work was created for Casey Reas's undergraduate course, Live Cinema through Creative Coding.

23. “Fractional noise” mountains (c. 1982), developed by Benoît Mandelbrot and Richard F. Voss at IBM, were a landmark in mathematical terrain synthesis.

24. Everest Pipkin generates barren flowerpot landscapes in Mirror Lake (2015), a poetic and mysterious browser experience.

25. In Jared Tarbell's classic Substrate (2003), simulated urban tectonics arise from elementary principles of accretion, branching, and feedback.

Additional Projects

• Memo Akten and Daniel Berio, Bozork Quest, 2013, scene generated with a fragment shader.
• Tom Beddard, Surface Detail, 2011, evolving fractal landscape.
• Tom Betts, British Countryside Generator, 2014, procedural world engine.
• Ian Cheng, Emissaries, 2015–2017, trilogy of evolving animated worlds.
• Char Davies, Osmose, 1995, interactive VR.
• Field.io, Interim Camp, 2009, generative software and film.
• Simon Geilfus, Muon glNext, 2014, landscape generation software.
• Chaim Gingold, Earth: A Primer, 2015, interactive book app.
• Beatrice Glow, Mannahatta VR: Envisioning Lenapeway, 2016, immersive visualization.
• Michel Gondry, Chemical Brothers “Star Guitar,” 2003, video clip.
• Vi Hart et al., Float, 2015, virtual reality game.
• Hello Games, No Man's Sky, 2016, multiplayer video game.
• Robert Hodgin, Audio-Generated Landscape, 2008, audio-generated landscape system.
• Robert Hodgin, Meander, 2020, procedural map generator.
• Anders Hoff, Isopleth, 2015, virtual landscape generator.
• Joanie Lemercier, La Montagne, 2016–2018, digital print on paper and projection.
• Jon McCormack, Morphogenesis Series, 2001–2004, computer model and prints on photo media.
• Joe McKay, Sunset Solitaire, 2007, software projection and performance.
• Vera Molnár, Variations St. Victoire, 1989–1996, silkscreen prints on canvas.
• Anastasia Opara, Procedural Lake Village, 2017, generative 3D landscape.
• Paolo Pedercini and Everest Pipkin, Lichenia, 2019, city building game.
• Planetside Software, Terragen, 2008, scenery generator software.
• Davide Quayola, Pleasant Places, 2015, digital paintings.
• Jonathan Zawada, Over Time, 2011, 3D models and oil on canvas.

Readings

1. Kate Compton, Joseph C. Osborn, and Michael Mateas, “Generative Methods” (paper presented at 4th Workshop on Procedural Content Generation in Games, Chania, Greece, May 2013).
2. Ian Cheng, “Worlding Raga: 2—What Is a World?” Ribbonfarm, Constructions in Magical Thinking (blog), March 5, 2019.
3. Philip Galanter, “Generative Art Theory,” in A Companion to Digital Art, ed. Christiane Paul (Hoboken, NJ: John Wiley & Sons, Inc., 2016), 146–175.
4. Robert Hodgin, “Default Title, Double Click to Edit” (lecture, Eyeo Festival, Minneapolis, MN, June 2014).
5. Jon McCormack et al., “Ten Questions Concerning Generative Computer Art,” Leonardo 47, no. 2 (April 2014): 135–141.
6. Paolo Pedercini, “SimCities and SimCrises” (lecture, 1st International City Gaming Conference, Rotterdam, Netherlands, 2017).