As i write this there's a 3 megapixel relic of the digital age sitting next to my keyboard. It's an Olympus Mju 300, a little all-weather (read: will resist the odd splash of water) compact that I bought around 2004, or thereabouts. It doesn't get much use now, although the fact that it's still here 7 years on says something about its appeal.
By modern standards it's no great shakes. The screen is tiny, there's a token (and all but useless) optical viewfinder and, of course, the pixel count is lower than the cameras built into most phones these days. But it's small(ish), the clamshell lens cover is clever and, provided you use it within its limitations, it's perfectly capable of turning out decent prints up to around A4.
I mention it because recently I had the opportunity to test half a dozen of its more modern counterparts for a brief magazine review. In the time since I bought the little Olympus things have moved on. Deeper menus, bigger screens and more pixels - a lot more pixels - are the most obvious changes. And yet... there was only one camera in this (hand picked) selection of current compacts that I'd spend my own money on.
That's not particularly impressive for nearly a decade of development. And it begs the question, why?
There are, as far as I'm concerned, three main issues:
1. Megapixel overload
The cameras I reviewed boasted between 10 and 14 million pixels. On paper that's enough to embarrass the expensive dSLRs that I use every day to earn my living. But of course not all pixels are created equal, particularly in cameras where they're crammed onto a tiny chip. Without exception, the 12 and 14 million pixel cameras' image quality was demonstrably worse than the best of the 10 Mp crop. I knew this would be the case, but I was shocked at how bad the results were, even at base ISO. Although I didn't get around to trying it, I'd put money on my 7 year old 3 Mp camera equalling or bettering the 14 Mp newcomers in an A4 print. That's progress?
Less is most definitely more, but the depressing reality is that camera manufacturers - with a few honourable exceptions - continue to chase more pixels. Truly a case of marketing over engineering - and in what universe do Facebook uploads need millions of pixels anyway? Bonkers.
2. Button stabbing madness
I'm old enough to remember when the first 35mm film SLRs started appearing with buttons instead of dials, back in the '80s. It was hailed (more accurately, marketed) as an advance. We had to suffer through a few years of cameras looking increasingly like pocket calculators with a lens grafted on the front before sanity returned and the wheel was quietly reinvented. Dials - or, with more complex cameras, buttons and dials together - rock.
Dials move, natch. And I suspect that makes them too expensive for most compacts, which increasingly look like (small) pocket calculators. If you're going to ask users to change settings via a menu accessed through a series of buttons, the UI becomes very important. And it's just not an area that many camera manufacturers get even close to right. Sorry, but a camera that forces me to press several buttons just to change (say) ISO or white balance isn't going to get my vote.
Funnily enough, the camera that emerged on top had both a well-designed menu system and a couple of concessions to old-school dial settings.
Nearly a decade of processor development, and yet we're still stuck with cameras that impose a significant delay between pressing the button and anything useful actually happening. Whether it's because there are more pixels to interrogate or some other arcane electronic engineering hurdle that I'm not aware of, the end result is the same. Unless your subject is anchored to the spot you're more likely to miss the shot than get it.
Before anyone accuses me of setting unrealistically high standards for cameras that are aimed at the mass market, I'm not just talking about bikes here. Most people use their cameras for people pictures. Some of the best people 'moments' are spontaneous. How are you going to capture them with a camera that doesn't take the shot when you ask it to?
You could reasonably infer that I'm not especially impressed, and you'd be right. What I see is a market driven by marketing rather than engineering, squeezed at the bottom end by camera phones and at the top by the new breed of EVIL cameras and cheap dSLRs. The result isn't pretty. It's certainly barely an improvement on what we had at the beginning of the last decade.
Oh, the camera that I'd actually buy? Canon's Powershot S95. It's a long way from perfect, but it stands head and shoulders above most of the competition.