Single Metrics

It happens all the time: someone will come come up with an interesting idea or theory that provides a simple answer by measuring only one thing. Larger sensors have shallower focus, smaller sensors allow better telecentricity; zoom lenses permit the exactly correct focal lengths, primes enforce discipline. Dynamic range, low noise, sharpness, contrast, wide, long: there's always some thought of the moment about what really makes a difference.

I'm a very technical photographer. I get hung up on the little things, I look for the small differences, and I try to figure out the setting that I'm missing. So I've been there. But as I become more experienced – read: cynical and jaded – I've realized that image quality isn't about solving a puzzle.

It's too tempting to want easy answers, or even to quest after elusive and esoteric ones. The reality is that it isn't a complicated question. Ultimately there's only one single metric that matters: is the photo satisfying?


Comments, questions, thoughts? You can find me on Twitter or via e-mail.

Protective Filters

The use of 'protective' filters – either UV-blocking or just clear glass – are one of those eternal internet arguments that people do instead of looking at photos. Naturally, I have a point of view about it: I don't see them having any actual use. I'm not about to say that anyone who disagrees with me is wrong – this is just my reasoning, and of course everyone else should ignore me and do whatever they're comfortable and happy with.

• Filters are the cheapest element in a lens. They're a high-margin item, so it's guaranteed that every step in the chain is making a healthy profit on it, and the materials and fabrication costs are going to be very low relative to the lens itself.

• By necessity, filters are the largest diameter glass in the lens, and are held in place only by narrow metal edges.

• In addition to their large span, filters are flat glass, rather than the domed shapes of the lens elements.

All of this adds up to a filter being the most fragile part of a lens. So to say "I dropped my lens, and the filter saved it" is a false syllogism. Just because the filter breaks on impact doesn't mean that the lens would have if it was absent. And just because all of the glass is intact, even if it's including the filter, doesn't mean that the impact didn't damage the internal mechanisms.

An analogy: Cars have a lot of safety systems built into them. Saying that a roll cage, airbag, or seatbelt saved a life could be a perfectly literal statement. Saying it about the bumper is not at all the same thing.

The other "protection" that a filter is supposed to provide is against scratches. I have talked to one person who had a small fleck of the surface coating flake off of a lens, but scratches and damage are exceptionally rare. Modern glass and coatings are very, very tough – including, incidentally, the ones on cheap filters. I once got my hands on a really cheap one – it was visibly off-colour and had a warped filter ring, brand new – and I tried to scratch it by hitting it with my keys. No luck. I eventually took my pocket knife to it, and still couldn't mark the front of it. If a no-name filter that couldn't have cost more than a dollar to produce can withstand that, then I don't see much reason to worry about the front element of a lens from a reputable manufacturer.

As a further example, I wear eyeglasses every moment that I'm awake, and have for twenty-four years. They've always been anti-reflective coated plastic, far softer than the glass in camera lenses. They're exposed to all kinds of abuse and neglect, worn in all weather, and I'll clean them more times in a day than I'll clean a camera lens in a month. Out of the half-dozen pairs of glasses that I've used in that time I can only think of one that I've scratched.

I do keep a hood on all of my lenses, which can provide both genuine impact protection and image quality benefits, but I rarely use lens caps. Two dozen lenses (or more) later, I've never seen a scratch or a fleck – and that includes on my used lenses, some of which have been older than I am.


Comments, questions, thoughts? You can find me on Twitter or via e-mail.

Colour Film

I have an ambivalent relationship with colour film.

I've recently spent a lot of time becoming familiar with the digital wonderchild D800, and it is a truly superb camera. It's effortless to use, and can do things that I wouldn't have thought possible. But now I've switched back to my hasselblad and zeiss rangefinder for a couple of weeks, and as it does any time I go back to a favourite, it feels like coming home.

I love the experience of using these cameras, and enjoy that they can't do some of the things that the digital ones do. Film really does have its own look, and it's one that I really like. There are 'filters' and 'plug-ins' available that are supposed to replicate some of that for the digikids; apparently there are techniques to make acrylic paint look like watercolour, but I don't see the point of that either.

But scanning colour film is one part of the processes that I do not enjoy.

When I started out with film I thought this would be easy: it has a fixed colour balance, so it should be consistent and straightforward. If only! When I'm at my best I will start the roll with a colour reference frame and set up the scan job for a consistent exposure and register the film base colour, and that helps. When I'm not at my best – like when I'm using my old light-leaking XA on a sunny day in New York City – then the results can be 'variable', to say the least. I can spend hours on a single frame and still come away with a result that maddeningly resists being good enough to be worth the effort.

Black and white is easy. It's also something that film is still superior to digital for. I just wish I didn't like colour as much as I do.


Comments, questions, thoughts? You can find me on Twitter or via e-mail.

Better Cameras, Part Three

Limitations aren't a bad thing for art or artists, although they can cripple work-for-hire photographers. The camera market and online photo-forum culture is very heavily driven by the idealized needs of the commercial photographer and the relentless march of More Better, but that doesn't mean that it has any bearing on creating actual art.

I believe very strongly that it's important to have different cameras, just as it's worthwhile to have different lenses. Working with different cameras, from pocketable to tripod-mounted, digital or film, is how I limit my choices and narrow my focus – figuratively speaking, of course.

My moving from generation to generation of digital cameras has been a process of removing technical limitations. I could print bigger, photograph in lower light, and have more maleable image files at every step. Removing restrictions can open up entirely new creative options and allow expression that simply wouldn't have been possible before. I can't advocate needless restrictions just for the sake of having some – everyone should be lucky enough to be able to experiment, find their own expression, and change as needed.

But being able to do anything isn't everything – and the people who are happy with what they do have generally found what they need already.


Comments, questions, thoughts? You can find me on Twitter or via e-mail.

Better Cameras, Part Two

I'm a big believer in working backwards: starting from the desired end result and then choosing the equipment that's suitable for it.

The photos that I'm most interested in generally have specific needs as they travel from being a digital file to a finished electronic image or print. Some need to be in black and white while others require colour; some have to be in a square crop while others are long skinny rectangles. Some photos require an immediate impact that can own a room, while others are best served by the more intimate and interactive experience of a hand-held print.

Of course, many photos aren't so temperamental. At that point it's a personal choice for what suits the purpose and vision the best. Projects naturally evolve and change, and the tools can be part of that. My ongoing 'time and motion' series started with my smallest film camera, the Olympus XA, moved on to a medium-format Fuji 645, and will probably involve my Hasselblad 6x6 next. And while I've picked an aspect ratio for them – 1:1.618 – I'm still not sure what size will be best for the prints, although I suspect it will be large.

Higher resolution digital and larger film formats can produce larger prints, and newer digital cameras generally provide better image quality than older ones. But that doesn't always matter, and there are times when I will choose the more limited tool specifically for the way it will impose itself on the process and results.


Comments, questions, thoughts? You can find me on Twitter or via e-mail.

Better Cameras, Part One

I don't believe in the idea of a "best" camera, and I'm weaning myself off of the word "better" as well. Cameras are certainly different, which is a good thing, as otherwise I wouldn't be able to justify owning so many of them. But to call one camera "better" than another is to impose a value judgement, and assumes needs or criteria that are far from universal. Sometimes quite large differences can completely fail to be significant.

I have a functional definition of "significant difference". If a difference is big enough that I would choose one something over another, then it's significant. If my goal is to produce finely detailed hand-holdable prints, typically from 6x8 to 8x10" sizes, then the resolution difference between my five megapixel Olympus E-1, built in 2003, and my shiny new thirty-six megapixel Nikon D800 isn't significant. If I control the light and choose the subject, which I typically do, then the other generational differences can go away as well. Both cameras will produce excellent, if not indistinguishable, results.

If all other factors stay the same, but I want to change the print to a finely detailed 16x20, then the resolution difference becomes significant. If the light goes down, or the subject moves, the other generational differences become huge. Working with the D800 does open a huge range of options to me, although that's also a subject for another day. Overwhelmingly, the latest-and-greatest isn't even stressed in conditions that weren't even possible five or ten years ago – although it's fair to point out that moving the goalposts doesn't mean that it's a new game.

It seems absurd to resist calling the D800 "better" than the E-1. In almost every measure it's a far superior camera. But my E-1 is still what I choose when I need a tough and quiet digital SLR, giving me the ability to take photos that I otherwise wouldn't have. And I also use it for other reasons that can't be measured or compared: there are times when I simply want the experience of using it, and there are times when I want the particular look and temperament that it provides to the photographs.

How can I call the D800 "better" than my E-1 – or F5, or XA, or Zeiss Ikon, or Hasselblad, or Fujis, or Panasonics – when I still chose to put it down and pick up something else?


Comments, questions, thoughts? You can find me on Twitter or via e-mail.

Left or Right

I really like this photo, and it's one of the few that I thought would work even as I was taking it. Perhaps that's why I don't have any variations on it. I wasn't planning on spending much time taking photos that day, so I was only carrying my point-and-shoot, and didn't linger.

It feels like there could be the beginning of a series here: the geometric experience of vast and deserted suburban streets. In my downtown neighbourhood it's rare to see a street this wide, and it would never be completely empty. Yet the evidence of people is irrefutable: someone built this road, and quite recently at that, and it has already seen enough traffic to wear away the markings. I find that contradiction fascinating, and at least as compelling as the opposing arrows.

I've done three different edits of this photo, starting from a clean copy of the image file each time. It's a way to take a fresh start but with the benefit of experience. This version was put through DxO Optics as well as Lightroom, with noticeably better results than working with Lightroom alone. Unfortunately for this photo, it's still not improved enough for me to be happy with it as a print. I like my little Canon S100, but it just doesn't hold up under close examination.


Comments, questions, thoughts? You can find me on Twitter or via e-mail.