It’s easy enough to understand why some people spend huge sums of money on wristwatches – aside from vices like vanity, avarice, and ego, there’s the whole “fine timepiece” marketing mystique, where watches are equated with jewelry, rather than treated as a functional piece of clothing, or a functional accessory, like a belt.
The manufacturers of “fine timepieces” tout a huge variety of snake-oil claims to promote their wares’ superiority and prices; one of the most common examples is timekeeping accuracy. The funny thing they don’t tell you, though, is that quartz watches – i.e. “electronic” watches – are all, for all practical purposes, about equally accurate. (We’re talking variations in single-digit parts per million, here.)
Now, it can’t be argued that with mechanical timepieces, not all are created equal; at the high end of things, you have watches that have surpassed mere watch-hood and become “chronometers”, with dozens of jewels and accuracy rivaling that of electronic timepieces. Does anyone really need this kind of accuracy – one second per week, and so on? It’s debatable, honestly; even back in the day when train conductors had specially-certified pocketwatches, the standards of accuracy were surprisingly low. Still, it’s an excuse to charge more money, which is all the motivation manufacturers need. It also raises two interesting questions: How bad, really, are low-end mechanical watches? And, of somewhat more widespread interest, how come a $35 digital wristwatch is always more accurate than a $350, or even $3500, computer? (My desktop PC loses around three minutes a week; the server that hosts the images for this site loses about twice that.)
The latter is a surprisingly simple difficult question to answer: It appears that, basically, nobody considers a computer’s real-time clock (i.e. the hardware clock) to be particularly important, so nobody ever invests in the parts and design to produce ones that are particularly accurate. With the widespread availability of network-based time references, and software to synchronize computers’ clocks, the impetus for manufacturers to make computers whose timekeeping is reasonably accurate is now largely nonexistent. It’s sad and pathetic, but that seems to be the way it is.
The mechanical watch question, by contrast, is surprisingly easy to answer. Behold, the least-ostentatious new mechanical wristwatch I could find last year:
That’s a Russian-made Pobeda, whose “2602” movement’s specifications are quite remarkably unimpressive: 15 jewels, and a “guaranteed” accuracy of -20 to +60 seconds per day. It was made in the last couple of years, but the movement traces its ancestry back to the mid-1930s, and nothing in it would be at all unfamiliar to any watchmaker from the past sixty or so years. It’s neither waterproof not shockproof; the case is brass and stainless steel.
I picked mine up on eBay from a Ukrainian dealer for the princely sum of $37.50, including shipping and a hideous leather band. It’s pretty much the least expensive new mechanical wristwatch you can buy, at least easily – the Chinese produce some cheap mechanical pocketwatches, and it’s rumored that a company in India makes some inexpensive wristwatches, but nobody seems to sell ’em online.
The Pobeda – whose name means “victory”, in Russian – has been one of my two everyday watches for about ten months, now. It keeps remarkably good time, for a mechanical watch: it’s consistently accurate to within about 90 seconds a week, or less than fifteen seconds a day. (It’s probably better than that, but the tiny second hand – which can’t be paused, or “hacked” – isn’t terribly easy to read.) Yeah, it’s far worse than a quartz watch, but let’s put it in perspective: that’s better than any of the computers in my house, and frankly, good enough for most people.
I mean, with the Pobeda, I know it’s only accurate to within a minute or two, unless I’ve just adjusted it, and that’s fine. I mean, time is fairly arbitrary, anyway: the “official” time is usually a few minutes off from what the local buses consider “official”, let alone what time my computer thinks it is, and those times usually differ from what time the thousand-and-one television stations and networks – who themselves don’t agree on what time it is! – seem to use.
I mean, in a world where Mythbusters starts at 7:58pm instead of 8:00pm, and the buses consistently say it’s 11:30am when my cellphone says it’s 11:28am, what’s “perfect accuracy” really worth, in real-world terms?