Electronic Design

Bob's Mailbox

Dear Mr. Pease,

Thanks for your timely, if somewhat desperate, reply to my letter of April 16th.  Some points of clarification are in order, and even a gratuitous confession or two of doubt.

Your assurance that the Slough of Despond is just around the corner for digital designers is based upon your assumption that we will inevitably sink into it when we try to run our clocks just a little bit faster than fast.  Optimistically, I have faith in all those heavy-hitter scientists who are shrinking our technology.  The true limit won't arrive until all components are in atomic sizes, and arranged in 3D matrices-sometime after our retirements, I presume.

Several more stages beyond "ultra" will come and go before digital engineers suffer analog constraints.  For example, I'm looking forward to the first invasion of digital wave-shaping circuitry into those arcane and untrammeled bastions of analog design:  Microwave and X-ray.

Let's take a quick shot at demonstrating my assertion about minimum energy.  When a digital circuit adds one more bit of width (extended throughout the system), then the new system can in fact handle twice as much information as the old system in any given time.  This additional bit, for example, could be a seventeenth after a previous sixteen, which amounts to a 6% incremental energy increase to the system.  We're not talking cost or even good taste here, just plain-vanilla information theory.  Now to make the same improvement to an analog circuit, one would have to double its signal-to-noise performance.  The usual unsubtle solution is to simply double the signal amplitude.  And, voila!, we have doubled the energy expenditure.  Q.E.D.

Digital speedometers give digital designers a bad image.  Numeric displays of time measurement, by watches and speedometers, leave out important information.  A proper speedometer display would be, of course, digital, because digital is the most accurate.  But the display should convey "how fast" by means of the relative position of a lighted indicator icon, a digital D'Arsonval meter that is stable and linear, if you will excuse the informality.  The word "Digital" does not impose digits on its designers, although too many of them think it does.  "Discrete" is a more exact term to distinguish our precise craft from the analog arts.

Once I had to indulge a client who insisted upon a D'Arsonval needle display, even when all the sophisticated real-time processor circuits inside his test device were digital.  We gave him a nice old fashioned bank of milliammeters, each one fed by a DAC.  A mere eight bits were sufficient to satisfy his analog illusion.

The atomic forbidden zones, as I understand them from college physics classes of long ago, are spherical shell-shaped regions around atomic nuclei in which NO electrons can EVER be located, although electrons can pass through.  Analog presumptions about a continuum of space are shaken by this phenomenon.

Suppose we consider a point on the periphery of a rolling wheel.  Its motion describes a cusp as seen by a stationary observer.  How long does that point remain in contact with a straight rail that is in tangential contact with the rolling wheel?

The analog engineer leaps to an illusory conclusion, "zero time," and almost immediately his finely honed skeptical mind cuts in, contradicting the obvious paradox, leaving him in a confused Newtonian babble about increments "taken to the limit," etc.

This real-world problem is easily comprehended by a mind trained in the digital craft.  There is a contact stress of the wheel against the rail that in the real world must be non-zero, because the condition is the definition of "contact."  Therefore, the resultant strain, however small, distorts the wheel such that a point remains in contact with the rail for a non-zero duration of time.  Now if we get clever and go optical, using shadow projections of a rolling wheel in stressless contact with the rail, we open another can of worms.  How can we measure an event of zero time?  When the engineering details of the experiment are considered, you can see the problem is truly insurmountable.  Analog is truly an illusion.

Organic perceptors of sound have either simple tympanic sensors that detect the presence or absence of vibrations, or tympanic/cochlear sensors that separately detect the presence or absence of vibrations at various frequencies.  Ears are inherently digital in the frequency domain, just like CDs are digital in the amplitude domain.  If Bach had only known.

The apparent dualistic nature of photons looks to me like the hint of a new physical theory just around the corner.  Photons are digital.  I KNOW they are digital, therefore their strange behavior indicates the presence of an unknown factor in the measurement.  Statistics isn't a very satisfying solution.

Are you soon going to hold forth on fractals?  It is a truly fascinating digital subject.

Peter R. Vokac, Digital engineer, Tucson, Ariz.

I'm sure glad we agree that ears are digital.  I have two.  (Maybe you do, too).  But you shouldn't assume that a guy who says "zero time" must be an analog engineer.  A lot of analog engineers understand the Real World better than digital guys.  Just listening to you confirms my opinion.-RAP

Dear Mr. Pease,

I have just finished reading your Pease Porridge column, "What's All This Spreadsheet Stuff, Anyhow?" from the August 20th issue.  I agree with your observations in this column and previous columns concerning the loss of the ability to think.  I have read a book that struck me as one you might enjoy reading; Technopoly:  The Surrender of Culture to Technology, by Neil Postman.  You and Mr. Postman address similar problems from different viewpoints.

David Curtin, Overland Park, Kans.

Thanks for your comments.  I'll have our librarian buy that book.-RAP

Dear Mr. Pease,

From time to time I have read your articles and have found them refreshingly insightful.  The wisdom of experience clearly shows through in your writing.  After reading the Pease Porridge in the August 20th issue, however, I felt you were not very balanced in your assessment.  While your comments were true in large measure, your omissions were blatant.

As I recall past articles, you have encouraged people to use their brains as God intended, i.e. to think.  In this article, you move dramatically against your own advice.

The theme of the article is spreadsheets are to blame for gross inaccuracies in calculated results.  But doesn't the real blame lie with the person who has stopped thinking due to sloppiness or laziness?

As if spreadsheets could really be "thought-annihilating," Michael Schrage and you blame the technology for people's own irresponsibility.  I have benefitted dramatically throughout my life from technological improvements, including spreadsheets.  I continually strive not to turn off my brain because of that technology.  What, or better, who is the real culprit.  Pease Porridge cold.

Lane O. Kagey, AST Research Inc., Irvine, Calif.

All I'm trying to do is counteract the mindless optimism of the spreadsheet peddlers who NEVER tell you to check your results.-RAP

Dear Mr. Pease,

As usual, I enjoyed your article on the July 23rd issue.  I recall hearing the Muntz stereo ads on the radio when I lived in the bay area in the 1960s.

Along with the design work I do here, I teach electronics at Cuesta College.  Your description of Muntzing a circuit reminds me of an attitude I try to instill in my students:  "The ideal design has zero parts."

I see too many "brute force" designs where the designer appears to have kept throwing parts at the design until it worked.  I prefer an elegant design, approaching zero parts.  One way I try to minimize parts is to move everything I can from hardware to software.

Harold Hallikainen, San Luis Obispo, Calif.

Didn't Albert Einstein say, "Everything should be as simple as possible, but not simpler?"  It's not trivial to say that software complexity is the right price to pay for hardware simplicity.-RAP

Dear Bob,

... What a helluva response to the Mr. X problem in the sampling of letters you put in the August 6th issue.  The excellent responses from EET engineers was heartening in that I realized I am not the only one having problems of recognition, and the age discrimination problem the first writer talked about.  I wish the two anonymous writers would have had the moxie to let the world know who they are.  And I totally disagree, as you do, with the guy from Minneapolis.

We ought to consider a few corollaries to the Mr. X problem one of which is indicated by one of the anonymous writers.  If, as apparently a majority of EET folks feel, we have a recognition problem, then why don't we as a group feel inclined enough to become part of the solution by identification, correspondence, and organization.

I would be more than pleased to correspond with other EETs about this problem.  But as long as we, as a group, remain generally anonymous in fear of our jobs or reputations, then nothing will be done.  Only through organization can we expect to solve the problems of apparent misrepresentation by our schools as to what an EET really is, where we fit into the engineering society, and what constitutes qualification.  As Benjamin Franklin still utters from his grave, "If we don't hang together, we, most assuredly, will hang separately!"

The second problem is accepting another simple fact of life.  Medical doctors, lawyers, and architects receive the fees they charge, gouge the public, and restrict those who can enter their profession simply because they have developed their professions over hundreds of years.  And in developing their professions over centuries, they have commanded the respect of their clients and developed for their clients true value for their product, like it or not.

Electrical engineering, by contrast, is rather immature.  Compared to doctors, lawyers, and architects, we electrical engineers are nothing more than brats.  As a group, we care only for ourselves, and not the profession.  As a rule, our professional societies are in business to perpetuate the myths and enhance their own bureaucracies.  Some how, some way, we need to enhance our customers' respect to the point where they value our services as they would value the services of their doctor or lawyer.  Only by increasing the customers' value can we expect our rates to increase and our salaries to grow.  And we must determine what that value is.

You see, our society is very malleable, like silly putty.  We expect our televisions, which come from abroad, to tell us what to eat, what to drink, what to drive, and what to think.  And through that august medium, as well as books and movies, we, as electrical engineers and scientists, have been continually depicted as self-serving nerds with no regard for anything other than our inventions.  We have permitted popular perceptions of the fruits of our minds, be it in radio, television, or nuclear power generation, to be damning, destructive, and dominating.  There is no value.  And why do they get away with perpetuating that persona of electrical people?  As my old friend Jack B. says, "Because they know they can!"

Chuck Albert, Engineering consultant, 700 Rosedale Dr., SE Smyrna, GA 30082

It will take us a while to earn respect for the Engineering Profession, let's keep working for it.  Part of the problem is we make things easy when they aren't, and vice versa.-RAP

 

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish