September 24, 1998

Why not the best?

Bill Schweber, Technical Editor

Lately, I've been seeing an increasing number of "best" lists: the 100 best films as judged by the American Film Institute, the 100 best novels as decided by the editorial board of Modern Library, the 10 best mutual funds, the 25 best hospitals, and the 20 best retirement areas, to name just a few.

Why do we see all these lists? As the editor of one financial-advice magazine bluntly puts it, "It grabs people's attention. It sells magazines." The fact that list makers can so neatly quantify difficult choices and alternatives apparently appeals to many people.

If only the life of a design engineer was so easy. Can you see being guided by articles entitled, "The 10 best op amps" or "The 10 top DRAM"? Of course not—and you know it. Your challenge is to decide what's best in your application, based on the criteria that you are using in that situation. In some designs, the driving force is a single dominant parameter, such as power, speed, precision, or that eternal constraint of price. You would be willing to yield on most of the other factors to get outstanding specifications in just that important one.

By contrast, in other designs, you have to carefully balance many factors to achieve the right combination of them. It's always a challenge to make those trade-offs, and that's where true engineering design expertise really counts. It gets even more complicated when you have to add in hard-to-assess factors, such as design risk, vendor credibility, and time to market.

Why is this issue important to you? EDN's role is to help you understand the trade-offs and the give-and-take dilemmas you face as you juggle a few primary performance factors plus a few less critical but still important ones. In the real world, there is no simple "best" answer. Instead, you have to make many legitimate choices, each bringing both some good and some bad. If we tell you there's a single right answer, we haven't done our jobs, and we're not helping you do yours, either.

Many years ago, before microprocessors ruled the earth, I took a graduate course in various techniques of optimization. Whenever a student would say, "The optimum solution is...," the professor would give us a verbal thwack on the head, reminding us that it is meaningless to say "optimum solution" without identifying it as optimum with respect to a stated parameter. It's a lesson I never forgot, and it's a lesson that you, as design engineers, live with every day.

You can reach Technical Editor Bill Schweber at 1-617-558-4484, fax 1-617-558-4470, or bill.schweber@cahners.com.

EDN Access | Feedback | Table of Contents |