But our pundits were right on in other cases. Nearly all agreed that enormous, graphics-enabling computing power would soon find its way to desktops. Some also predicted the advent of compression algorithms that would make it easier to view and share video and other applications on the desktop. Others predicted the ubiquitousness of multimedia computers in the home. And several foretold the success of Win dows NT as a workstation platform.
In this, the first issue of the new millennium, it seemed appropriate to look backward-and forward, as well. We tracked down several of our experts, who gamely agreed not only to comment on their eight-year-old predictions, but to offer their forecasts for the next 5 to 10 years. Their opinions, then and now, follow:
1992 · By 1994, virtual reality will become more practical due to ordinary glasses being equipped with small lasers that will impinge directly on the retina of the eye. Also by 1994, we'll start seeing the rapid deployment of display systems with built-in cameras that will be used for teleconferencing and video email.
2000 · In 1992, the promise and hope of virtual reality seemed within our grasp. But the difficulties have turned out to be greater than we thought: The user's physical interface is still limited to displays, keyboards, joysticks, and, recently, usable voice recognition. Based on what we've learned over the past eight years, I'll have to extend my vision of a truly immersive (even semi-immersive) virtual world to 10 or even 20 years into the future.
In the meantime, computer graphics will continue to make extraordinary strides, delivering photorealistic images on commonplace computers. Real-time as well as stored media information will be presented to us on high-resolution multi-format screens ranging in size from portable telephones or PDAs to multi-monitor and large projector devices, in rich color, with smooth frame rates and perfectly synchronized sound.
1992 · Probably the most significant thing we are going to see in the next five years is the emergence of Windows NT as an operating environment that gives us all the capability we have in a Unix workstation today.
In 10 years, I think that, in terms of CAD/CAM, what we're going to see is a far more intelligent modeling capability that has embedded in it some understanding of the manufacturing processes. For example, suppose you were an automobile designer designing sheet metal for a car. I think that future systems will understand that the designer is working with sheet metal, and they'll understand the properties of sheet metal, in terms of how far it can be bent or how sharp a radius it can be bent on.
2000· Every major CAD vendor now has an NT version of its software. But the incorporation of manufacturing constraints into CAD software hasn't gotten as far as I had ex pected, though sheet-metal forming is pretty common and well integrated in core CAD software packages. In the next 5 to 10 years, I expect we'll see more CAD software that understands material properties and manufacturing pro cesses of the sort I expected eight years ago. And within five years, we should see Internet protocols integrated with most CAD applications.
Computer makers are going to find it increasingly costly to boost clock rates. Increased computing power will come from applying multiple processors to applications. The CAD industry is woefully unprepared for this hardware trend. The outfit that figures out how to take advantage of future multiprocessor architectures will gain market share at the expense of its competitors.
1992 · I think that certainly over the next five years, multimedia will become ubiquitous. By that time, you will not buy a workstation that does not have multimedia capabilities.
In 10 years, a couple of things are going to happen. For instance, I think voice input will be much more common at that time, and I think that virtual reality will be accomplished without the need for restrictive head gear.
2000 · Certainly sound and CD capability are pretty standard with PCs now, and video is frequently available. Voice input, however, still isn't very common-although it is available. Non-HMD VR displays exist now-products like the immersive desk from Fakespace, and the lenticular lens-like screen from DTI provide effective display environments.
Certainly what we didn't call and what undoubtedly will be the environment of the next decade is the Internet. Another development that changes the rules is the pervasiveness of extremely low-cost PCs.
1992 · Low-cost, 3D rapid prototyping systems [will] become available....We'll also see the traditional CAD/CAM marketing strategy of a pyramid that divides products into high-range, mid-range, and low-end categories become totally bankrupt. Instead, CAD/CAM companies must map into their users' communications needs: global, site, and niche ap plication integration.
2000 · Prices of low-cost 3D rapid prototyping systems have dropped significantly with the introduction of "desktop" systems under $50,000, but the industry needs a killer app (like personalized action figures printed on demand at McDonald's!) to get prices to lower levels.
With regard to the pricing of CAD applications, many high-end vendors like PTC and SDRC are seeing slower growth in CAD and are turning to PDM and data management as their best bet for meeting business expectations. Dassault and Unigraphics have bought into the mid-range. At the low end, cheap 2D and 3D CAD is becoming a commodity, and once-proud vendors like Cadkey are struggling. Emerging trends to deliver and support CAD software via the Web will add further pressure.
Global integration, site integration, and niche application integration, as predicted above, but enabled by the Web, are the most exciting aspects of today's technology.
1992 · In the next five years, I see an increase in the level of what Nicholas Negroponte called "intimacy;" that is, the broadening of the bandwidth between the user and the computer to un precedented levels, the extreme that we visualize right now being immersive virtual reality.
2000 · My careful prediction from 1992 is coming to pass, but is still probably another five years away. I believe that five years from now, there will be a wide variety of small devices interconnected with and without wires, making life more interesting, if not simpler, for everyone. These devices mostly won't be referred to as computers, but will simply be part of your doors, walls, appliances, clothing, car, and toys.
And there won't be any major engineering projects that are not run with the help of extranets.
1992 · Probably the most exciting development will be a new process for the way engineers do their work. So far all we've had are tools to automate the way engineers have worked for the last 200 years. The most exciting thing will be when we get to a new way of doing engineering, a new way of doing engineering in teams.
2000 · Today, the Web has begun to change the world in so many ways, it's hard to imagine where it will end. In manufacturing, globalization, collaboration, build-to-order, and upside-down customer-driven processes will evolve as the enormous potential of the Web is better understood and utilized. As for CAD/CAM/CAE, 3D modeling will predominate and physical prototypes will only be used for product testing where they are government mandated. As for 2D, it'll be as quaint as green eyeshades and drawing boards are today.