Image/Work: Notes on Warburton’s RBGFAQ

Deborah Levitt

Bookmarks, Switches, and Levers

In a new video essay about the history and future of computer graphics, RGBFAQ, Alan Warburton explores the layers that make up what he calls the “exploded image.” Digital images, he suggests, are always multiple. GGI are intersections of processes of construction and composition, from modeling, rigging, and animating to light effects, volumetrics, mo-cap, and more. In a beautifully animated sequence about render elements, Warburton flips the layers of the exploded image like a deck of cards or sheaf of photos, rotating and spinning them through virtual space, then lining them up in their heterogenous array to show us what they look like to the animators, designers, engineers, and developers that work with them—and to demonstrate how fine and precise their controls are. Cycling through examples of beauty, matte, geometry, and utility elements, Warburton demonstrates how these images function as sites for manipulation, intervention, and revision through clever exaggerations and shifts in palette. There are visualizations of processes of visualization, like the XYZ RBG motif that pictures a red, blue, and green XYZ axis floating in the gridded interior of a cube. The sequence performs reflexively, thematizing its own conditions of production.

RGBFAQ is Warburton’s fourth video essay; it is also his longest and most ambitious. it follows Goodbye Uncanny Valley (2017), Spectacle, Spam, Simulation (2016), and Fairytales of Motion (20??) Across these essays, and in much of his work in experimental CGI as well, Warburton stages an argument for his own practice of research-creation and critical making. And even where his own practice is not the focus, practice itself, particularly in the rituals and repertoires of bodies at work, are central to Warburton’s project.

In RGBFAQ, computer generated images are work spaces. The exploded image, Warburton explains, is “a type of hyperimage – an image with bookmarks and switches and levers embedded into it.”  The image appears as factory, as operating theater, as office machine, as theater of operations. This view responds—sometimes explicitly and always implicitly—to the contemporary tendency in media theory to stress the disjunction between the micro and macro spaces and times of computational processes and the mammalian timespace of the human sensorium. The black box is the common emblem for this computational opacity, and no doubt this phenomenon presents a growing challenge to perception and understanding. Warburton approaches the conundrum of perceptual access from a different angle. RGBFQ focuses on sites of production (rather than on reception or consumption). Warburton engages computational image-making through what Gaston Bachelard, in his works on the philosophy of science, called phenomenotechnique. That is, Warburton does not pursue perception and understanding as separate and autonomous categories, but rather works from within the exchanges between instruments, practices, and the knowledges they reflect and produce. 

Warburton reflects on the labor of computer graphics and CGI as a series of experiments involving gestural sequences and materials (as image infrastructures). The exploded image, in particular, emerges from the hacks and work arounds designed to minimize the computing power needed for complex renders. Its modular protocol develops from the need to be able to edit images without expensive and time intensive re-renderings of an entire scene. Each layer of a computer generated image is produced by parameters that isolate an aspect of the image that can be extracted, manipulated separately, and then returned to the bigger picture. Contingent, shaped by budget pressures and the infrastructural pushback of the apparatus in the form of artifacts, glitches, errors, latency, and what Neta Alexander calls the “existential condition of buffering,” the hyperimage, for Warburton is  “a taxonomy of things we can know from an image, as well as a toolbox of operations that can be performed on an image,” where things we can know and operations that can be performed are inseparable.

Recursions: Data and Display

Funny thing is, the problematic of perceptual access to data is what launches Warburton’s history in the first place. In the opening chapter, Warburton tells the story of Edward Zajac, the mathematician at Bell Labs, who in 1963 created the first CGI movie. As Warburton narrates, Zajac, who was working on the mathematics for satellite orbits, was exhausted by scanning the rows and rows of numerical equations printed onto the reams of papers that held satellite data, and so turned his focus to “how to intuitively represent massive amounts of spatial data.” Warburton in turn works on visualizing the process that leads from Zajac’s wireframes through the multilayered photoreal CGI/computer graphics of today. According to Warburton, the strange loops in the development of computational data and display, bring us full spiral from the moment after Zajac’s solution when computer graphics and computer vision parted ways to their contemporary reunification as “dual apparatus of the computational image.” Machine learning is trained on artificial data sets. Data is simulated and simulations are sources of data. It’s turtles—or maybe camels—all the way down.

At least as far as humans are concerned. While I’m not certain that we have passed through the looking glass of simulation once and for all, there’s no doubt that our fate is bound to the fate of the image, our futures to its futures. Giorgio Agamben once wrote that the human is not defined by rationality or language or technics. Instead, and most essentially he suggests, the human is the moviegoing animal. But perhaps it might be better to say—drawing on the multivalence of the term—that beyond representation and simulation the human is the animal that renders.

Deborah Levitt is Assistant Professor of Culture and Media Studies at Eugene Lang College, The New School. She is a media historian and theorist interested in film and digital media including CGI, VR, and AR and has published on Giorgio Agamben, media and biopolitics, and animation theory. She lives in New York.