Hyperfocal Length, Depth of Field, Circle of Confusions

by igmillichip on Sat Oct 16, 2010 12:27 am

Hyperfocal length is a very important value for certain types of photography.

You need to know the ‘Acceptable’ Circles of Confusion for your camera.
The Circles of Confusion depend on a number of things, but one item is the size of the film.
You have a Nikon D60 and that is the same of many digital SLRs…..high quality, large sensor but NOT actually 35mm film size (the Nikon D3 is different for example).

Thus an acceptable Circle of Confusion for your camera is approx 0.02mm (same as mine)….a Nikon D3 (and similar formats) is 0.03mm but compacts are much smaller.

The size of the circle of confusion is important in that it tells you the size of a cone of light focused upon a film at which, any larger the cone, the point of light will appear as a circle instead of a focused point of light….ie out-of-focus.
If the end photograph is to be printed on larger and larger print, then the focusing must become more critical (as them circles of confusions will also be enlarged)

CALCULATIONS
So, now take that Circle of Confusion for your camera (0.02) and multiply it by the f-number at which you are going to shoot.

Eg 0.02 x 3.6 = 0.072 We will call that {N.CoC}

Now, Take the focal length (in mm) of your lens and square it. Eg 35mm x 35mm = 1225mm

Divide that number by {N.CoC} to get 1225/0.072 = 17013mm = ~17metres = Hyperfocal distance.

Now, that is based upon conveniently viewing a 10 inch print at a comfortable foot away.

Now, what that means is that if you focus your lens to that distance (and not the object) then, roughly, things from half that hyperfocal distance to infinity should be acceptably sharp.

If you try to increase the aperture number to get a greater depth of field and reduced hyperfocal length then you run the risk of lens diffraction occurring (which is negate any attempt to get sharp images). For your camera, f16 is probably the limit to which you can close your aperture down further before diffraction effects start to show (ie don’t got to higher f-numbers with a 1.6 scaled DSLR)

I’ve tried not to go into too much theory there…hope that helps.
Sometimes the theory does not come to fruition in practice, as there is some ‘experiential’ input into all of this anyway.

I’ll recap the calculation for a lens of focal length 35mm, and the f-number set to f3.6.
If a camera if focused at 8 metres, then anything between 5.5 and 15 metres should be in acceptable focus.

So….to reinforce the theory….
If

f = focal length of lens,
N = Aperture number (f-number),
And CoC = circle of confusion (use 0.02 for most digital 35mm SLRs; don’t bother with compact digitals; use 0.03 for Nikon D3 and some of the newer high-end Canon pro DSLRs, and higher still for medium and large format film)

Then
Hyperfocal Distance (HD’) is approx = (f * f)/{N * CoC}

In the example given, Aperture, f- number, (N) = 3.6; focal length (f) = 35mm; and CoC = 0.02

So, HD’ = {35 * 35}/{3.6 * 0.02 }
= 1225/0.072 = 17013 mm = 17 metres.

(in ideal situation, the Hyperfocal Distance is actually a slightly different formula…but there is an allowable approximation that I have included. If we did the ideal then we would get 16,978mm rather than 17,013mm ie approx 17 metres !!!)

Hypernerdy or what??

Once you have calculated your Hyperfocal Distance for a given lens and given f-number, then you can also calculate the ‘depth of field’ when focused at any given distance…ie the closest and further distance that objects will be in acceptable focus.

These approximated calculations do not work when dealing with macro photography especially if using extension systems (eg bellows or extension tubes).

In practice, much of the above is not overtly relevant. If doing a course in photography then some courses may expect you know it; if using large format cameras then you’ll have to know it; if using a 35mm then you’ll start to notice things within the photograph that just don’t look or expected.

Good photography is not about taking photographs of ‘lens-test-charts’, that is for large format photographers to do and lens manufacturers to carry out.

Fortunately, many of the lenses that are used by many people on Digital cameras are actually crap……if the lenses used were the very high resolving lenses then all of the flaws of the digital sensor would become very very noticeable.

You’d see too much ‘Aliases’ on the image (fine repeating patterns would be a nightmare to look at); you’d get spurious spikes in regions of sharpness (which would look artificial).

When I attach my high-end lenses from my old Nikon film cameras to digital, the images are a bit duff. So I’d hate to see a proper lens (like Schneider pro-lenses) being used with digital cameras. If, however, the camera has a very strong anti-alias filter built in (and my digital does not have a strong anti-alias filter) then those problems may not be noticeable.

Digital SLRs also have a ‘built-in- softfocus system….by way of a problem with large f-numbers and low resolving power of digital cameras. (in fact what is happening is that the contrast is becoming very low and will approach zero in some cases)

Use apertures above f8 (eg f11 upto f32 or whatever) and you’ll get softened images (low contrast images really). Take care when the ‘alias’ pixels start to kick in though…..ie that ‘digital’ look (ever seen a newsreader wearing a fine tweed jacket?….and you wish he hadn’t done so!!)

Digital cameras (as will all cameras) will get a theoretical crisper and fuller contrast focusing at low aperture numbers; but at low aperture numbers any flaws within lens manufacture start to show and the photographer needs to be very very critical with the focusing as there is very little depth-of-field.

At a photography course this will be dealt with as ‘MTF’ or Modular Transfer Functions; the theory is very complex; BUT…… at the practical level….this is dealt with as a ‘rule of thumb’ advice.

ian