At college, or wherever the last place you studied mathematics was, you were probably told to show your work and not to jump to an analysis or a conclusion without explanation. The way that you tackled a problem was as important as whether you actually got the correct answer.
But it’s not always true that is the best way to proceed. My father started driving a vehicle rather late in life and mercifully took himself out of the driving seat before his abilities were really impaired. He went the formal driving school route in search of his license. During the first lesson, the instructor tried to explain the workings of the internal combustion engine and how speed and gears were all related to one another. My father cut him short and told him that he didn’t care what was in front and below him, he just wanted to drive.
From an engineering perspective, we would probably insist that not understanding the relationship between engine speed, gearing and road speed would make it very difficult to understand how to use a gear box, but he wanted to learn to drive just by rote.
We can see a similar pattern in broadcast engineering over the years in knowing, or not wanting to know, the internal workings of products. There were days when you had to completely understand the circuit operation of a piece of equipment, or else you would not be able to tackle any repairs on it; repairing products down to the board level for all the equipment in the studio and transmitter was the norm, not the exception. Today that is no longer true for most equipment and most repairs are either at a higher, modular level, or equipment is completely chucked away.
What many engineers do not realize is how that also relates to the components that are used to achieve the results we want to see. When I came into the industry, a sync pulse generator was a rack-mounted unit (yes, with tubes!) and was unreliable, temperature-sensitive and always had a redundant backup. Then Ferranti produced an IC that was most of an SPG (it missed some broad pulses in the 50Hz version, as I remember), and there was a loud bell that went off in my head about what linear ICs might do in the future.
The broadcast industry has not been a prime target over the years for specialized ICs. Sure, there have been thousands of devices that have been used, but the main targets for the products have been in the much larger professional or consumer device arena. And that is probably not going to change in the foreseeable future, so we need to be careful not to ignore some advances in technology and technique that directly affect the way we design and use equipment.
During a recent trip to the San Francisco Bay area, it was a pleasure to be involved in giving out awards to a number of semiconductor companies. At one of the last stops, I was invited into a lab to see a prototype chip in operation; I was flabbergasted at what I was looking at. The IC was a software-configurable processor. In the demonstration, it was crunching an HDTV signal into other formats and sizes — in effect doing things that just a few years ago required rack-size boxes with mil-spec ICs, and a lot of them. Here was a $100 IC, with a very small number of peripheral components, that could power a box that broadcast customers would pay tens of thousands for with a standards converter or video manipulation title attached to it.
So, if you see people at NAB wearing a badge with the company name Stretch, Inc., stop and ask them what they can do for your application. But don’t ask them what is under the hood — you really don’t need to know.
Paul McGoldrick is an industry consultant based on the West Coast.