Grant Stevens

Real-time procedurally generated computer graphics with sound

Audio consultant: James Peter Brown

Supported by UNSW Faculty of Arts, Design and Architecture

Single white words fade in and out against a background of slowly changing block colour. The words are from a custom database of more than 1300 words that describe the full spectrum of human emotions. Each word is algorithmically sequenced based on its emotional proximity to the preceding word. These words link with conjunctions and adjectives to create an endless, uninterrupted textual flow. The text is accompanied by a synth soundtrack responding to the emotional tenor of the words on screen. The pacing of text and sound also modulates between rapid and slow.

Real-time procedurally generated computer graphics with sound; audio samples by Ilya Basov

Installation with custom furniture and programmed LED lighting sequence

Lines of white text gently fade in and out on screen against a slowly changing skyscape. The phrases are algorithmically selected and sequenced from a database combining text from mindfulness meditation, body scanning, self-affirmation, guided imagery, mantra, and positive psychology. The soundtrack is peaceful and always changing.

Real-time procedurally generated computer graphics with sound, endless duration

A camera floats in a desert valley at dawn. Clouds gradually form and dissipate, sometimes becoming overcast. Occasionally there is fog. The hum of a string ensemble hints at the possibility of narrative, but action and resolution never arrive. The camera is in constant motion, programmed with AI to cinematically pan, tilt, and crane. Sometimes the sky becomes the sole focus, while at other times, arid foothills and distant mountains fill the frame.

Procedurally generated computer graphics with sound, multiple display formats

Originally commissioned by Buxton Contemporary for the Light Source Commissions series

Assisted by Pat Younis

An animated juvenile deer is programmed with artificial intelligence. As it wanders through a computer-generated redwood forest, it is followed by an AI camera that tracks its movements to create a continuous POV shot rendering in real-time.

Below the mountains and beyond the desert, a river runs through a valley of forests and grasslands, towards an ocean (2020)

Procedurally generated computer graphics with sound, multiple display formats (1-3 video channels), endless duration

Originally commissioned by UTS (The University of Technology Sydney)

Assisted by Max Warlond

Seventy virtual cameras continuously pan and track through a large computer-generated landscape featuring forests, deserts, mountains, lakes, oceans, and grasslands. Natural cycles of days, nights, seasons, tides, and the weather are programmed to create endless changes in real time. The camera movement is slow, but time is sped up. One year of running time represents 100 years. Over many years of running time, geological features are eroded and uplifted; trees grow and decay.

Watch interview here.

Procedurally generated computer graphics with sound, multiple display formats (1-3 video channels), endless duration

An artificially intelligent camera roams endlessly through an idealised computer-generated forest

The Forest live streamed for 28 days in October-November 2020 as part of the Institute of Modern Art’s Making Art Work program.

Procedurally generated computer graphics with sound, endless duration

A virtual camera rotates around a computer-generated mountain. The scene changes according to the time of day and the weather, which are programmed using artificial intelligence.

Assisted by Nicola Best

4K video with sound, 5 min 29 sec (mix and additional sound production: Beau Cassidy and Jonathan Wilson)

Mantras and phrases borrowed from mindfulness meditation, corporate training, cognitive therapy, online dating, and job applications mix together on screen, accompanied by an uplifting soundtrack.

4K video with sound (multiple display formats, 1-3 video channels), 8 min

A male narrator describes facts about the universe, while the lifecycle of the universe is digitally visualised.

HD video with sound, 6 min 15 sec

A male narrator describes facts about the Sun, while a virtual camera moves gradually closer to a digitally animated solar mass.

HD video with 5.1 surround sound (multiple display formats, 1-5 video channels)

A virtual camera floats across a panorama of endlessly generating digital clouds. Imagery and audio are both digitally generated to create infinitely variable patterns and loops that avoid obvious progression and climax.

archival ink on archival paper, 11 prints, each 82.5 x 55 cm

synchronised three-channel digital video, 7 min 17 sec

As an uneasy soundtrack reverberates and distorts, shots of rainforest undergrowth become increasingly menacing and disorienting.

digital video, 3 min 31 sec

A female voice guides a meditation while an abstract animated form progressively morphs from geometric to irregular on screen. The tone shifts from relaxation to seduction and, eventually, to hints of mild contempt.

digital video, 3 min 15 sec

A series of white words and phrases fade in and out as a virtual camera flies towards the abstract horizon line of a computer-generated dawn. The soundtrack, text and camera movements shifting from optimistic to unsteady.

Just Dawn was commissioned for the It’s Timely exhibition at the Blacktown Arts Centre in April 2014 (curated by Paul Howard and Gary Carsley), and is a response to two speeches that former Australian Prime Minister Gough Whitlam delivered in Blacktown in 1972 and 1974.

synchronised four-channel digital video with sound, 11 min 19 sec

Four virtual cameras take different paths through an animated three-dimensional scene. The scene is comprised of forty-four clusters of slowly orbiting white text, each referring to a different topic that has been sourced online. The topics are diverse. Some relate to spirituality, others science, popular culture, food and various experiences of contemporary urban life.

digital video, 3 min 3 sec

As a computer-generated waterfall water gushes, text borrowed from a popular talk show host’s self-help advice fade in and out graphically down the screen. The animated phrases increase in tempo, and the sounds of the waterfall begin to overwhelm the tender stock music.

lenticular prints mounted on dibond, 6 panels, each 50 x 30 cm

digital video with sound, 5 min 30 sec (sound by Josh Mannis)

White words and phrases fade quickly in and out amongst small bursting dots of colour. Set to an energetic, synthesised soundtrack, the animated text mixes an internal monologue with dialogue from overheard conversations at an unspecified crowded social event.

Commissioned by the Museum of Contemporary Art Australia 2012 to acknowledge the MCA's Building Donors

digital video, 5 min 7 sec

A female voice softly recites physical and psychological associations of aura colours. On screen, individual words fade in and out rhythmically amid a field of swirling and morphing colours. The animated words correlate with the words being spoken, but not every word is displayed.

digital video, 5 min 9 sec

The 'camera' zooms through and around a constellation of one-thousand words (ten words, each repeated ten a hundred times) suggesting an atomised mantra. The words pulse independently as they orbit an invisible nucleus. Airy, synthesised soundtrack.

digital video, 4 min 11 sec

Phrases offering a heartfelt account of a failed romance fade in and out on screen. At first they appear at a readable pace, but, as the quantity of words increases, it becomes more difficult to read them all. Sad piano soundtrack.

digital video, 18 min 17 sec

A couple in the throws of breaking up converse on a couch. They are filmed in conventional shot/reverse-shot style. However, sound and image fall out of synch. Sometimes they appear to mouth each other's words.

digital video, 3 min 5 sec

Texts sourced from dream-description websites are combined into an improbable narrative. The words float against an animated cloud background that subtly changes colour. Slightly anxious stock music.

lambda print, custom cabinet, speakers, 80 x 178.5 x 40cm, audio track 4 min 57 sec

digital video, 7 min 33 sec

Phrases from personal profiles on social-networking websites are formatted as a mandala, which slowly rotates and changes colour to a meditative soundtrack.

lambda print, custom cabinet, car stereo, 75 x 197 x 39 cm, sound by Rex Goh, 23 min 37 sec

digital video, 3 min 27 sec

The text of a long-distance love letter is dispersed throughout the cosmos, accompanied by romantic piano musak. The word-clusters can be read in different orders, but the sentiment remains the same.

two-channel digital video, 4 min & 10 min 57 sec

Text from a travel blog types our across the left-hand screen, a paragraph at a time. A computer-generated image of a tropical sunset is slowly animated on the right-hand screen. Atmospheric corporate groove stock music.

digital video, 3 min 14 sec 

In The Switch, plots from two teen films of different genres are spliced into one. The text types out across the screen a line at a time, accompanied by a swelling orchestral soundtrack.

digital video, 2 min 53 sec

There are two competing narratives, one written, one spoken. Both constructed entirely from cliches and mixed metaphors. The words flashing on screen and those heard in the authoritative broadcaster-style voiceover are sometimes the same, and sometimes different.

digital video, 3 min 3 sec

Phrases from film previews are combined to suggest a larger, albeit unlikely, plot. The words approach us from the black background, accompanied by driving stock music.