Integration in electronics — what does that even mean? Whatever it is, people are talking about it more and more, so when I was asked to write something about integrated analog, I was intrigued.
Naively, I thought I would just reach into the Blogging Hat to pull out some thoughts from a lifetime spent integrating (OK, sometimes disintegrating) electronic systems. After all, it’s a chance to talk about the design processes we use to create electronic systems, instead of the subtle magicks involved in chasing signals around and submitting them to cruel and unusual transfer functions. Have those design processes changed beyond all recognition, or have we actually been this way before?
I started out thinking that a plan for this blog would be a good idea. But no plan survives contact with the enemy. In this case, the enemy is the pressure to contribute regularly, coupled with the unanticipated twists and turns created by focused feedback from the army of Planet Analog and EDN readers. So I’ve decided not to follow a preordained path through a set of issues. Instead, I’ll light a few torches and see where the crowd carries them, throwing in my own views whenever the ground becomes stable enough to put down a soapbox without fear that it will sink and topple.
My perspective is colored (I wanted to write “coloured,” but my iPad changed it) by several factors. First, I’ve been doing electronics for all my adult life and for a while before that. So in the last 40-odd years of holding a soldering iron or waggling a mouse, I’ve seen a lot of things go down (and go bang). Second, I’m an analog guy (guess what my iPad does there), and the burden of that endangered-speciesdom seems to press more each year. Integrating analog without understanding it seems like an awesomely bad idea to me.
Finally, I work for a company, Cypress Semiconductor, that’s working diligently to be in the lead pack of suppliers leveraging the integration trend. Oh, sorry, I just pressed the make-it-sound-like-marketing button. But you get the picture.
Professionally, I lean toward the belief that integration will and should happen, rather than a wait-and-see agnosticism. That probably makes me a punching bag for all the things that people are not going to like about the integration thing. I won’t use quote marks for the word, but since we still don’t really know what it means, we’ll have to tolerate our collectively imprecise use of it.
At the start, I asked whether we’d actually been this way before, and I think we have. As a boy hobbyist taking my first steps in the late 1960s, I thought a bright future awaited in which we would replace tens (perhaps hundreds) of transistors and passives with a single IC. Today, we are all agog at the thought of replacing tens (perhaps hundreds) of ordinary ICs and passives with a more highly integrated device, for which we haven’t even yet got a good acronym.
There are plenty of questions queuing up for an answer, but I’ll close this first epistle with some quotations from the editorial of the September 1968 edition of the UK magazine Practical Electronics. Launching a series introducing the home constructor to the systematic use of the integrated circuit, the editorial observed:
No enterprising constructor will wish to stand aside from the mainstream of emerging technologies. Even on the home constructor scale definite economic advantages will soon materialize from the outpourings of the micro-electronic plants. The cost of a “one-off” integrated circuit will become less than the equivalent discrete components. Nor is it fanciful to envisage IC’s [sic] at “give-away” prices in due time.
The magazine staff agonized over how to structure the projects using a very early operational amplifier. They presciently realized that “something more than the mere fitting of a ‘black box’ into a circuit was required if the problem was to be properly tackled.” The role of the articles was to “provide an introduction to system designing and to building ultimately on a larger scale — and this is how we are likely to make the most profitable use of ICs [sic] in the future.”
So I’m sure we’ll get many opportunities in 2013 and beyond to debate the editorial’s assertion that the projects “will also bring out the fact that circuit design will not become a redundant art even when microelectronic devices take over the major role in electronic equipment.”