Now, with digital SLRs, autofocus and high resolution sensors, you might wonder how things have changed?

Well, now you can get an app for your phone that will work out the appropriate distance setting…

…but, as I often ask of people on our courses, have you actually looked at how sharp your photos are when you trust to the magic numbers?

Remember to treat everything you read about the ‘best’ ways to do something in photography with due scepticism. Try things for yourself and let experiment triumph over dogma…

I’ll start with a brief overview.

Hyperfocal focusing – some background

First up, I know that almost any maths frightens off a lot of photographers, so I’m going to avoid too many numbers, wherever possible. I’ll include more detailed references for those wanting to follow up on the equations.

If you want to test what I’m saying, just take your camera outside, set it up on a tripod, follow some simple procedures for focusing and take some photos.

Then, after looking at the results, decide what you think is relevant, when you might want to use hyperfocal techniques and when not to.

Anyway, what is hyperfocal focusing, and why do so many people suggest its use?

When you focus your camera lens on an object, what’s behind the object and what’s in front of it are not focused so clearly. The depth of this zone of sharpness (front to back) is thicker for a small aperture (say f/11) than a wide aperture (say f/2.8). The precise thickness of this zone depends on what -you- decide is acceptably sharp. It’s also known as ‘depth of field’ or DOF

Lenses used to have ‘Depth of field lines’ inscribed on them to give an idea of how much DOF you got with each aperture setting. The four below are from my ‘junk box’ collection at 55mm, 70-150mm, 300mm and 500mm

It turns out that there are ways of calculating the thickness of this zone, but these depend on all kinds of factors and assumptions, such as how big a print you want to make, how good we think peoples’ eyes are at resolving detail and not just the more concrete numbers such as lens aperture setting, focal length and size of the camera sensor (or film) you are using.

The key idea is that there is an ‘optimal’ or hyperfocal distance (HFD) that for any lens/sensor and setting your focus to it will give you the thickest zone of sharpness that includes ‘infinity’.

Ah yes, infinity…

If you set your lens at infinity, then there is no concept of ‘further’ or ‘behind’ (further than infinity is still infinity)

You can think of just a ‘near point’ in this case, where focus becomes sharp enough for your needs, as you recede from a camera with the lens set at infinity. Everything beyond this near point is ‘sharp enough’.

Note that there are actually two definitions of Hyperfocal distance – very similar in practice. If you want to look at the history of this idea, the description on WP is quite comprehensive (moderate maths warning ;-).

Hyperfocal focusing assumes that if you focus at infinity, then you are in some way wasting some of that rear sharp zone, so focusing closer lets you use up some of it, and gain a closer ‘sharp enough’ point. You are trying to maximise depth of field and still include infinity.

In real life, there are no infinitely distant objects, so we equate ‘far away’ with infinity – just how far this is depends on your focal length (and sensor size too, but I’m going to keep this simple and just stick to FF/35mm/’FX’ for my experiments)

There are all kinds of calculators, charts and even phone apps that will crunch the numbers and give you a hyperfocal distance setting.

One example I used during my testing was this one at Nikonians. You enter a few numbers and the magic distance pops up.

What concerns me is the arbitrary precision that many of these calculators provide.

When you get an HFD value of 32.45 feet or 9.89 metres, it’s all too easy for people to assume that we’re talking of something that is accurate to the centimetre.

I see this approach all over the place, in printer and lens reviews, where there is often the unstated view that more precise numbers equate with greater authority/relevance for the data.

It’s this spurious numerical precision that can easily lull people into accepting what might be a helpful general technique as more of a law, written in stone. (WP article about False Precision)

Now, I know that there are photographers far happier in dealing with numbers, specifications and charts than exploring the personal/artistic/creative (i.e. vague and ill-defined) aspects of photography.

Even so, I’d still suggest that if you are going to use a technique such as hyperfocal focusing, then you should at least do a few experiments and realise the importance of the concept of ‘acceptable sharpness’.

It’s this term, that appears at the heart of all the definitions and equations that exposes the essentially subjective nature behind all those very precise numbers.

With ubiquitous presence of autofocus (AF), the detailed lens depth of field settings you used to get inscribed on the lens barrel are increasingly rare, so it actually becomes quite tricky to set a hyperfocal distance manually for many lenses. If you want to experiment with a particular HFD table or calculator, find something in your scene at the HFD and focus on that.

I should note that some people just assume that you should focus a third of the way into your scene, and that will give a suitable depth of field – this is almost always wrong, and invariably gives inferior results to using hyperfocal focusing. A while ago however, I found a different simple approach that fits very well with my own style of landscape work.

An alternative to HFD

First up I want to acknowledge my debt to the writings of Harold M. Merklinger of Halifax, Nova Scotia, Canada.

His books concerned with focusing view cameras have inspired my further experimentation with tilt/shift lenses on my dSLR and article on ‘how to focus with tilt‘. One problem (for some) is that he doesn’t shy away from including calculations and deriving equations for his various findings. I believe that this has unfortunately contributed to the lack of wider appreciation for some of his observations and suggestions.

In an elegantly simple demonstration (Depth of Field Revisited) he shows that:

By focusing your lens at infinity, the smallest object resolved in your image will effectively have the same width as the focal length divided by the aperture of your lens.

As ever, the details are a bit more complex than this, so do read the article and also see the links at the end of this one if you want more info about just why this works.

So, for a 90mm lens at f/10, this size is 90mm divided by 10, or 9mm.

For a 24mm lens at f/11 this is only 2.2mm (24 ÷ 11) or less than a tenth of an inch.

What do these numbers mean?

If I’m using a 24mm lens at f/11, focused at infinity, then 1 metre from the lens, I can resolve detail of a few millimetres, just as at 100 metres away I can resolve a few millimetres. Note that this doesn’t mean I can’t see things less than this size, just that they won’t be sharp.

Of course, with a real lens and camera, there is no way that I can really get a few millimetres resolution at 100 metres, but it does mean that I can get maximal sharpness for distant mountains, and still resolve blades of grass pretty close to the camera.

But – couldn’t I get even better sharpness of that nearby grass by using hyperfocal focusing?

Yes I could, but the distant mountains would lose detail.

In my large landscape prints I want lots of texture and sharpness in the background. If I’ve something in the foreground, then I generally don’t mind it not being quite so sharp.

Anyway, time to look at some examples (and please do try this for yourself!)

Some real life testing

Do remember that these are 21MP images reduced to web size, and the detail views are 100% crops with no more than default sharpening (RAW files processed in Adobe camera raw). More importantly, you’re seeing JPEG files in a web browser, so don’t get too picky over precise details (trying it yourself is always best).

The first set is taken with a Canon TS-E 24mm lens (no tilt or shift) at f/11. This is the street outside my house.

At f/11 the focal length divided by the aperture is 24/11, or just over 2mm

With all these images, move your mouse over the image to see a version with the focus set at infinity.