Since first gaining widespread acceptance well over a decade ago, CAD workstations have revolutionized industrial design, changing just about everything apart from the activity at the heart of the process: product conceptualization.

Most product ideas still emerge first as sketches drawn on envelopes, napkins, sketchbooks, or whatever happens to be at hand when the muse speaks. Then, as the idea is pushed forward into more advanced stages of development, a representation of it typically must be modeled on-line to allow the refinements and modifications necessary to satisfy engineering and manufacturing requirements. Given the inherent inefficiency of building a model only to rebuild it in another medium, there should be a better way to launch the design process.

THE HIGH GROUND
"Designs are the results of free play in the right hemisphere of the brain," explains Lisa Krohn, an instructor at Pasadena's Art Center College of Design. "And, unfortunately, computers have historically proved to be rigid in a way that's antithetical to these intuitive, creative meanderings. As a result, they've stifled the very part of the thought process most vital to conceptualization. Before computers can play a meaningful role in helping people shape and refine their ideas, they'll first have to show they can meet us on our terms."

Judging by what's going on at Art Center, generally regarded as the top school in graphics and industrial design in the U.S., there's reason to believe that some progress is already being made. In the school's computer graphics laboratory, a leading-edge facility that includes 35 Silicon Graphics desktop systems and workstations, important advances in designer-friendly interfaces are already in evidence. "Over the past couple of years, I've noticed that more and more of the students are spending less and less time working out their ideas on paper before moving to a computer," says Rob Hennigar, chair of the school's Computer Graphics Department. "In fact, we have a number of students now who seem to be doing all their concept work right on the computer. That's something you never saw before."

Marty Smith, chair of Art Center's Industrial Design Department, sees it as the beginning of a sea of change in design, driven primarily by the emergence of interfaces that allow designers to create digitized models of their concepts simply by sketching them. "In a way, " he says, "we've always taught our students to tap into their creativity through the drawing process. And, at last, interfaces fluid enough to let them create computer models in just that way are becoming available."

Interfaces notwithstanding, the advantages of using computer-based tools to develop concepts have been apparent for some time. The most familiar argument for creating 3D geometries as early as possible in the design process is the opportunity to reduce the time it takes to get a product to market, given that 3D models can be tested and refined for manufacturing even as the design itself continues to evolve. "When you're developing a design on a computer, you have more tools for testing your ideas, and you have more ways for quickly getting your ideas to the point where you can honestly evaluate them," Hennigar points out. "That means," he says, "without investing too much time or effort, you can decide whether to develop an idea further or reject it. And, generally speaking, the more ideas you can develop and test before committing to a design, the better your chance of getting a satisfactory end result." There's also the advantage of being able to edit a design instead of starting over from scratch whenever it becomes apparent that a portion of the design is flawed. In many ways, this is similar to the difference that led people to make the jump from typewriters to word processing systems some years ago.

ON THE MARCH
In combination with improved interfaces, these arguments make a compelling case for on-line design conceptualization. Smith says, "But up until a couple of years ago, the applications available for industrial design were actually better suited to design visualization than to design conceptualization. But now we can also say that there's some software that really works for conceptualization."

The difference became obvious earlier this year, Smith explains, in the course of a project sponsored by BMW that challenged students to develop ideas for a two-passenger sports car combining the thrill of a motorcycle with the safety of an automobile. At BMW's request, the students were directed to work exclusively with computer-based tools.

"To be honest, I was astonished at the quality of the sketches produced by the students," says Smith. "They had all the fluidity, emotion, and impact of conventional sketches. In fact, if anything, they were better because the computer provided a richer palette of colors to work with."

THE CHALLENGE AHEAD
Still, for all the advantages of on-line conceptualization, the challenge remains to develop interfaces that are more natural and intuitive for designers. Even the best of the currently available solutions doesn't completely capture the sketch metaphor that's so vital to designers.

"I'd imagine in the future we'll see systems that take advantage of flat-screen monitors you can place on a table facing up at you from the same angle as a sheet of paper," predicts Smith. "With the systems we have today, you have to look up 90 degrees from the digital pad you're sketching on to see the image you're producing. It's difficult to align the two movements, and it introduces an uncomfortable level of disassociation from the natural activity."

At the most recent Siggraph convention, Disney Imagineering demonstrated a system that approximates just such an effect by reflecting screen images off a glass drawing board situated in front of an illustrator. In all likelihood, other innovations intended to ease hand/eye coordination for on-line designers and illustrators will soon surface.

Clearly, the form such interfaces will take can only be imagined. But one thing is certain: the approaches that succeed will be those that are least at odds with the "hands-on" way in which people traditionally have grappled with industrial design problems. Until such interfaces emerge, the CAD revolution will not be complete.