A few years ago, I upgraded from a trusty but ageing 12MP DSLR to the then-latest model, a camera with a 24MP sensor. My choice wasn’t driven by a desire to have a camera with more megapixels, more to be able to benefit from the slew of advantages offered by the current generation of cameras.

At that point this meant HD video, better focusing performance, a superior LCD screen and Wi-Fi among other things. The fact that my new camera would come with a 24MP sensor was just part of the deal.

One of my first tests was with a simple 50mm lens, and I remember it for one reason: I hadn’t expected to be so disappointed. Once stopped down it did pretty well but when used wide-open it just wasn’t very good at all.

It wasn’t a focusing issue and my lens wasn’t faulty – I just hadn’t previously had the opportunity to scrutinise an image captured with it to the same extent.

With an alternative lens I didn't have any issues, but at least here, I could see that my upgraded camera was essentially giving me worse results. Nothing about that lens had changed, of course, but all of a sudden it was difficult to view it as favourably as I had done before.

Information overload

Between details from manufacturers, technical tests, hands-on reviews, full-size image samples and observations from end users, we’re now informed to an unprecedented level as to the performance of a piece of kit.

We want to make the right decision so that we end up with something that matches our expectations, and most of us are prepared to put the work into finding out as much as we can to ensure this is the case. The problem is, it’s difficult to appreciate when exactly we’re being over-informed, and when the information we’re considering is irrelevant to our own specific situation.

A lens that’s capable of producing sharp results on one body may well behave different elsewhere. Between sample variation, slight misalignment of elements from poor handling (or damage during transit), focusing problems, camera compatibility, and, of course, the nature of the camera with which you partner the lens, two people can form a very different impression of the same lens.

How you use the lens is another factor to throw into the above pool of variables. Someone who habitually uses a wideangle lens at a small aperture for landscapes may praise its sharpness throughout the frame, never realising that it behaves far less impressively when used wide open. Similarly, a DSLR user that only ever uses live view may well see different results to someone that relies entirely on the separate AF sensor that comes into play when using the viewfinder.

Manufacturers like to draw our attention to MTF charts of a particular lens, but these only apply to us to a limited degree. The lens manufacturer has no idea what camera we’ll be using and how, what kind of anti-aliasing filter it has (if it has one at all) or whether we plan on using images straight out of the camera or interpolating Raw files with a third-party program.

These charts may be based on real-world measurements or on computed ones. They aren’t detailed enough to cover performances at all apertures, focal lengths or focusing distances; instead, they typically provide an idea of performance where they are likely to be used. As Nikon itself states, “MTF charts should only be used as a starting point when comparing and purchasing a lens. ”

What is a “good” lens?

All of this should go some way to highlighting why giving a lens – or indeed, a camera – an overall score is problematic. Assuming testing conditions are identical it’s an easy and quick way to understand whether one lens is considered to be superior to another, but we’re only likely to know for sure how happy we'll be with it once start using it ourselves.

After all, the final say on image quality is down to the human observer. Quality is something we ourselves define, not an intrinsic property of a lens.

A lens with its fair share of residual aberrations may be considered to be only an average performer in isolation, but if you’re using it with a camera that has the ability to automatically correct these, you’re likely to be much more satisfied than if you were to use it with a model that does not. Again, nothing has changed about the lens, it’s just you're being more pragmatic by taking the body into account.

A lens might not be considered entirely sharp, and may exhibit corner softeness and vignetting, but for the portrait photographer these features may be desirable rather than offputting. It may be a dream when paired with a full-frame body that has a modest pixel count but may be underwhelming when used with an APS-C body with a more populated sensor.

Of course, the more commonly heard side of this argument is that focusing on how sharp a lens is won't make you a better photographer in the same way that developing a better understanding of light, or factors that influence depth of field, might. Clearly one is easier than the other, but everyone has a right to delve into lens geekery as far as they want. It's just probably worthwhile to remember that the next time you buy a well-regarded lens, there are a number of reasons why you may not form the same impression that someone else did. Or, perhaps, you’ll like it more than you imagined.

2017-6-2 03:00