Lost In Space
by Andy Turudic
Anyone who has traveled extensively can easily attest to finding unsigned roads as part of supposedly easy directions, particularly main thoroughfares. Is eliminating signage, or clear direction, a way to save money, a scheme to confuse GPS-equipped invaders, or simply a message that outsiders are not welcome? Perhaps the reason is that familiarity fails to raise the flag of necessity in the administrators charged with putting up the signage and they are rewarded for such behavior by not having received any complaints from the locals.
With the advent of a new community come the repeat problems created by those new-on-the-job humans that are charged with its administration. In the virtual world, the car we drive is our computer, though we are cursed with the unusual phenomenon that our virtual 405 hp Z06 Corvette has the relative power, and value, of a Geo Metro just as the last payment comes up. Our highways are the URLs, sideroads are links inside a website, and the items we seek are the buttons we press to get PDFs, application notes, samples, or perhaps on almost rare occasion, information on contacting a real human being. We rely upon the marketers and webmasters to give us the signage that gets us where we want to be, hoping that they themselves recently drove the route once or twice in the mind's eye of the customer or client. Unfortunately, it's not always the case.
In recent times, "keeping my axe sharp," I found myself thrown back a few decades and actually doing a mixed-signal board design -- a consumer product, no less -- and seeking out components that would do the job and bring the design down within budgetary constraints. Ignoring the nickel-and-diming of every part, it actually was quite fun to do a serious board-level design again and was very illuminating with regard to the drivel some semiconductor marketers put out to direct prospective customers as to the location where applicable devices could be found or the spins used to describe devices' functionality. This lesson in futility was a valuable refresher for me in marketing components: if they don't market themselves, no design-in can be had unless each customer is personally attended to. If they are over-sold, or -hyped, it's a dead end and time to try another, more credible route. Yes, if you're a >$1 M account sometimes you get the attention of, and clarification from, a supplier, but not as a startup or consultant unless you've been a really squeaky wheel.
In my search for the avenue that would lead to an inexpensive, yet functional, design, I found distributors who were not afraid of publishing standard prices, allowing me to quickly weed out others that wanted social security number, firstborn's middle initial, and selection of an 18 character password, requiring at least one of those characters to be traceable in lineage directly back to the Rosetta Stone, before one could proceed with seeing if they even carried the part, let alone obtain a price. This Sherlock Holmes-esque marketing technique lends the specter of gnomes in a back room trying to piece together a design based upon component information requests. If I had to "call for quote" from a distributor, to me it likely meant the part was being passed through from the factory with no value added, they were having allocation or introduction problems, or that they were fat and dumb with the type of staff that adheres to a policy of returning a simple pricing message at least three days after a request is left on a rollover to voicemail. Did they get their parts onto my BOM? Nope. I made a point, after finding virtual responsiveness, and finding budgetary pricing to be reasonable, to prefer building the BOM from two distributors and only went outside the box with a couple of strange, sensor-type, components that were not on their line cards. If I was unhappy with the standard pricing on an occasional part, a quick phone call got me to someone I could try to pummel the price down with. These were the places I would go to find components, or the places that would link me to where I could get the device data. The design came in under budget and was a fully functional paper tiger in a couple of months.
Unfortunately, some of the places I was sent from the distributors' linecards were websites that had a "products" tab, and then a page full of just part numbers, underlined with hotlinks. No product description or category -- just numbers. Other sites only had an applications' taxonomy, with no part types or descriptions. Apparently some marketing VP decided that there was nothing else in the world but DVD players, MP3 players, cell phones, and PCs. This was familiar to me from over a decade ago when I sought a data sheet for the first 20-Mbyte floptical drive for an AWG I wanted to build; I kept being asked what PC it was for and could not get anyone in North America to send me a sheet for designing the hardware interface and firmware driver. All of these places had the road, but no sign identifying that I had gotten to the right place, or if they did have a sign, it quickly gave the impression of "dead end." I could spend a lot of time looking for my destination, or I could move on, with the manufacturer's only hope being that someone else, a magazine article, or a website like analogZONE would direct me, on a functionality or parametric basis, back to the right place.
The bottom line is that if a business has the fantasy of playing in the neighborhood of high volume consumer products, at least providing a quickly responsive budgetary price, ideally capable of being sorted in a parametric table, and understand that designers are trying to hit short product windows and that they work after 4 PM and weekends -- price is the first criterion, and verification of adequate functionality and performance is a close second. If you don't want to deal with the little guy, send him to a distributor with a well oiled website -- little guys sometimes wind up as chief designers at companies that would make a marketing VP salivate and they may come back with direct levels of unit volume. And above all, if it's good enough in a good-enough market, don't think you'll be successful fetching a price premium of 4x in a market that will garner you the joy of competing for 2% of the total units sold.
With all these lessons learned, I can now find what I'm looking for during
any board level design and get the job done quickly and without too much
waste. No doubt many of you have also established a least-energy algorithm
for locating components, favorite distributors, chip suppliers and web portals.
With my recent refresher as a board designer, I'd like to think I'll be
a more responsive semiconductor marketer as a result.
Contact the author