Redefining the World 

Facing off against the blue screen of death

Technology has a habit of sneaking up on you and making great leaps and strides when you're not paying attention. These changes are usually small and relatively insignificant — like, say, shrinking an MP3 file from 4.6 megabytes to 4.3.

Sometimes, though, what happens is literally ground-breaking, a moment when something is accomplished that has never been done before, when what has been accomplished can become a definitive before-and-after in that particular field. This is what is known as a paradigm shift, and it redefines the world around it.

Russell Hilleke, associate professor of physics, and John Peeples, professor of electrical engineering — both of whom teach at The Citadel — have been involved in these moments. Although you may have heard of one accomplishment and have practically no idea about the other, the importance of them collectively, and of countless others, helps to create an interesting commentary about the nature of research. Although I interviewed them on separate days and times, the number of similarities between their answers was surprising and gave some insight into the personal rewards that technological innovators get from their work.

Dr. Hilleke explained that the best thing about being an experimental physicist is that "the instruments are like toys," and that when you get an idea for something and you work on it long enough, your main focus and drive is that initiative. While he was working at Argonne National Laboratory in Chicago in the mid 1980s, he and a team of researchers built a device to measure the chemical and magnetic depth of solids without actually interfering with those solids themselves.

"When we first started developing the reflectometer, we could think of no reason why it shouldn't work," he said, his enthusiasm for the subject evident. After roughly two years of development, they were ready to try getting it to operate. "You're plugging away that whole time, flying blind, really, but following your intuition. And that first time we ran it, and we got it to work, well," he grinned with remembered pleasure, "that feeling was like Christmas morning."

He went on to say that the device was almost like looking through a prism at light patterns, but then quickly amended himself. "No, it was exactly like looking through a prism, except instead of light, you were seeing neutrons," Hilleke says. "The results weren't just what we expected. They were better than what we'd expected."

His team's success with this went beyond personal satisfaction, however; a short while later, IBM approached them to develop a similar device, this time to measure the depth of polymers for use in computer chips.

Which brings us to Dr. Peeples, who worked for NCR in the early '90s. "We'd been working with Intel to develop faster processors, but the biggest problem we encountered was the heat distribution issue," he told me. "Other companies, like Sun Microsystems, DEC, etc., were making chips with faster processing power, and we wanted to figure out a way to increase our speed." Working with Intel, they started to make strides in that area, creating a patent portfolio of 17 different designs.

Eventually, he and a small group of researchers left the company and, in early 1996, formed their own, Kryotech. "We had been primarily working with larger systems, servers and mainframes, but we wanted to focus our efforts on personal systems, to achieve the Four Minute Mile of computing — the one gigahertz computer." The one gigahertz computer had seemed an impossible goal to achieve; the faster a processor was, the more heat it generated. Eventually, without proper cooling, that chip overheats and cracks, rendering the computer inoperable. Trying to get past the barrier from MHz to GHz was more than just a benchmark. The way they figured out how to do that was by building a refrigeration unit into the system itself, attaching a cooling unit right onto the face of the processor. Working long days and living on their savings, they toiled away on their project, eventually getting some help from (ironically enough) Intel's chief rival, AMD.

Finally, in 1998, they could feel they were getting closer. Peeples says, "We'd get the chips, put them in the computer, start it up, and it would look like it was about to boot when we'd get the blue screen" he recounted ruefully. "All we could do was to keep trying."

Then, one Thursday in April, it happened. "In this field, the big joke is that nothing ever happens during regular hours. You usually get results at midnight on a Friday, or working all the way through the weekend," Peeples says. "The following Monday, there was a trade show in New York, and AMD had said they had 'something big' to unveil. So when we installed the chip, turned on the computer, watched it boot up, and then the Windows banner appeared on the screen at 4:30 in the afternoon, then we restarted and it booted up again, we were pleased beyond all expectations."

Aside from some quick drama — the head engineer accidentally cracked the chip, and they had to scramble for a replacement that worked before getting the system to the show — they managed to make it work and get it to keep working.

Looking back, both men had good things to say about their experiences. "You don't get many moments where you feel as though you've made an actual advance in technology," Dr. Peeples told me.

Dr. Hilleke was equally insightful. "I'd describe it by saying that you feel as if the group you're working with are the only people in the world who understand how important this is. Einstein once said that you never get to the point where a breakthrough is routine," he further observed. "I agree with that wholeheartedly."


Subscribe to this thread:

Add a comment

Classified Listings

Powered by Foundation   © Copyright 2016, Charleston City Paper   RSS