CG Revolution
Issue: Volume 35 Issue 6: (Oct/Nov 2012)

CG Revolution

Pixar toy teapots

Pixar RenderMan

Newell’s teapot drawing

Joaquin Kino Gil’s teapot (modeled in Wavefront, later rendered in Maya)

It’s difficult to pin down the exact date or event that spawned interactive computer graphics—the so-called big bang of our industry. Was it when Ivan Sutherland showed off the Sketchpad interactive graphics software he had developed at MIT? When William Fetter created takeoff and landing studies with wireframe aircraft at Boeing? Or when he placed a digital human inside a digital cockpit to test control reachability? Or perhaps when General Motors used a proto-CAD system from IBM to control a milling machine via computer? All these events occurred in 1963, and perhaps none served as the spark that ignited the CG industry, but there’s no doubt that each acted as a catalyst that placed us firmly on the path to the digital graphics era.

Shortly after those events, there were other significant developments, among them: the first computer graphics department was established at the University of Utah, SIGGRAPH began, and Martin Newell created the infamous CG teapot, which became a computer graphics icon used to demonstrate new methodologies. In terms of technology, we were introduced to Bezier curves, Gouraud shading, Phong shading, Z-buffer, and image and texture mapping. Despite the fact that CG was still a niche market driven by a select few, innovation was happening and driving the technology forward.

The “world” of computer graphics at this time was indeed a very small one. Nevertheless, it was a field occupied by some dedicated pioneers with the passion to create and innovate. Computers were becoming a thing of the present, rather than that of the future, and some—like Jim Blinn, Carl Machover, Ed Catmull, and others—believed the future included interactive CG.

“I was driven by the idea that I could integrate a numeral representation of an object with a graphical representation, and that by manipulating the graphical representation I would manipulate the underlying numerical representation,” Ivan Sutherland had stated so long ago about his Sketchpad software. And he was not alone in that regard.

Soon there were a number of math and science students who had turned to computer science only to find nirvana in the subset of computer graphics. Suddenly all those 1s and 0s began to paint a picture. “Computer graphics conjured up a rich palette of visionary ideas, of a seemingly infinite range of possibilities, grand concepts with the promise of big payoffs. It was hard not to imagine all that data that was being digitized for the first time, having coordinates assigned to it so it could be displayed on a screen or plotted out, illuminating the underlying relationships within the data,” stated Randall Stickrod about those early days. He would later turn the “Computer Graphics Newsletter” started by Joel and N’omi Orr into Computer Graphics World magazine. (Read the Editor’s Note on pg. 2 for more on the history of CG and CGW.)

Back in those early days of the industry, most of the developments in computer graphics occurred at universities, government labs, or at large aerospace or automotive facilities. And, they needed supercomputer strength to power them—eventually coming from Silicon Graphics machines before the PC revolution in the 1990s. Innovation began to occur rapidly, and the number of companies developing products grew from “a handful” to well over a hundred in early 1980. While computer graphics milestones were achieved in films such as TRON, most of the big developments originated within the CAD industry, where, as Stickrod explained, “the productivity gains associated with CAD systems were immediately obvious and compelling, prompting large-scale investment in these tools and technologies.”

Indeed, CG visual effects required a great deal of effort and expense. Scenes featuring the stained-glass knight (the first 3D character) in Young Sherlock Holmes, the water tentacle in The Abyss, the skeletons in Total Recall, the shape-shifting T-1000 in Terminator 2, virtual reality in The Lawnmower Man, and the dinosaurs in Jurassic Park were novel and appreciated by theater-goers.

After Toy Story, the first 3D computer-animated full-length feature film, debuted in the mid-1990s, it set off a spark in Hollywood, leading to CG features from PDI (Antz) and another from Pixar (A Bug’s Life). The stage was set, and CG infiltrated the silver screen with one CG-infused blockbuster after another. Today, the top 30-plus grossing films (worldwide) contain cutting-edge CGI, including the stereo 3D masterpiece Avatar, which holds the number one spot.

We have witnessed the introduction of photoreal CG characters and their maturation to where they not only act alongside their human counterparts, but also assume a leading role. We now have digital replicas of athletes and actors that are controlled in real-time gaming environments. And we have smart characters that will test concepts and products reliably and safely. We have seen studios tackle and perfect untold CG challenges, from skin and hair, to water and cloth. We have experienced them controlling motions and emotions without missing a beat. What was yesterday’s biggest CG challenge quickly becomes today’s norm—until a new challenge comes along.

CGI has evolved from images that have a plastic look to those that are indistinguishable from reality. Today, artists can explore aesthetics anywhere along that plane.

Over the years, we have witnessed the rise of many hardware and software companies, and their demise. Competitors have become partners, dominant players have vanished, newcomers have become the new leaders. Technologies have done likewise. While CG has become pervasive, there continues to be innovation in this realm, as those working in this industry continue to seek (and conquer) new frontiers.

“It’s hard to say anything about the progress of computer graphics beyond, ‘Wow, it sure has gotten better, more realistic, cheaper, and more available.’ But that is the basis of the paradigm shift from being an exotic laboratory phenomenon, to becoming an expensive professional tool where, in 1995, a computer-rendered Homer Simpson says, ‘I feel like I’m wasting a fortune just standing here,’ to the case now when computer graphics is the cheapo way to make images instead of using expensive, real-world sets and actors,” says Blinn. “But, let’s face it, the good old days were when it was incredibly difficult to get even a simple line drawing onto film. Now, an army of just plain folks can all indulge their imaginations and publish the results to the world via the Internet. I like that world better.”

In the subsequent pages, we first look at some memorable covers from the archives of CGW. Next, we revisit some industry milestones covered in the pages of our magazine. Although there are too many to show each one of them, we have attempted to highlight a number of events that raised the technical bar in CGI—a few sweet treats as we pass the teapot.

Karen Moltenbrey is the chief editor of Computer Graphics World.

The University of Utah is believed to have been the first to exhibit the following: (from left) texture with highlights (1975), texture with reflection mapping (1975), highlight simulation (1975), and bump mapping (1977).

The Revered Teapot

CG innovation has become synonymous with the iconic SIGGRAPH teapot ever since Martin Newell from the pioneering graphics program at the University of Utah introduced the virtual object to the community in the mid-1970s. Over the years, researchers have used the 3D computer model as a benchmark for image synthesis programs demonstrating new technology.

Most people in the computer graphics realm are familiar with the teapot, at least the digital version. However, there is an actual teapot that Newell used as a reference in his quest to develop a free, standard 3D model (which was difficult to produce at the time), and that one now resides at the Computer History Museum in Mountain View, California.

So why a teapot? As the story goes, Newell’s wife suggested the object while the two were having tea. And her suggestion, from a technical standpoint was “spot on” for a number of reasons cited over the years: It is round, contains saddle points, has a genus greater than zero because of the hole in the handle, can project a shadow onto itself, and looks reasonable when displayed without a complex surface texture.

In 2006, Peter Shirley, adjunct professor at the University of Utah, paid homage to this model in his SIGGRAPH Talk “The Teapot Through the Ages,” discussing how “this simple geometric model became a graphics icon” and showing some of the images that have poured forth over the years. We have seen wireframe versions, solid models, teapots with procedural textures and teapots that have been raytraced, ones that are furred and others that are flat shaded, some with reflections and refractions, and some with displacement shaders—there’s even one that was displayed on a high-dynamic range display with a fluid dynamics simulation of steam coming out the spout. What’s more, each year at SIGGRAPH, Pixar hands out hundreds of tiny wind-up teapots to collectors.

During the mid-1990s, the Stanford Bunny hopped onto the scene. Developed by Greg Turk and Marc Levoy while exploring range scans, the object comprises nearly 70,000 triangles and is better suited to illustrate certain concepts, such as radiosity and self-reflection, than the teapot.

“It’s been really fun seeing the teapot showing up in so many places. Why, it’s even a primitive element in many graphics libraries, up there with spheres, cones, and cylinders! (Maybe that’s one reason it shows up so much),” says CG pioneer Jim Blinn. “Of course, having a common object to test rendering algorithms on is useful so that you can see how differences in lighting and texturing appear, unaffected by differences in shape. But the teapot as modeled by Martin Newell is also an inherently beautiful object. I still enjoy looking at it, and look forward to seeing what people do with it in the future.”