Work in progress

What's an aperture? The word means “opening”, but in photography it's the aperture in the middle of the lens that lets light through. To control the amount of light, nearly all lenses allow the size of the aperture to be changed. This has a couple of interesting effects.

Controlling exposure

It turns out that the amount of light hitting a unit area of the sensor (or film) is dependent on the relationship between the focal length and the size of the aperture. Apertures are nearly always circular, so we can talk about the diameter of the aperture as the focal length divided by a number, say f/4 (aperture diameter is a quarter of the focal length), where f is the standard symbol for the focal length.

And up to a point, that's all that you really need to know about apertures. It's certainly by far the most important thing.

Depth of field

One of the properties of a lens is that it creates a sharp image only for objects of a certain distance, the focus. Objects that are not at that distance will be more or less unsharp, depending on the distance from the focus.

From diary:

One article in particular was titled What is equivalence and why should I care?. Good question, at least the first half. But it explains: “Equivalence, at its most simple, is a way of comparing different formats (sensor sizes) on a common basis.”. Well, not exactly a dictionary definition, but when I say that my Leica Summilux 25 mm f/1.4 is equivalent to a 50 mm f/1.4 Summilux on a full-frame camera, I'm using that definition. About the only thing I can fault in the description is that they don't go into aspect ratio. Here a comparison for a Micro Four Thirds camera with the 25 mm lens and a full-frame camera with a 50 mm lens:

Sensor format:		4/3		FT
Focal length:		25.00 mm		50.00 mm
Horizontal FOV:		38.17°		39.60°
Diagonal FOV:		46.81°		46.79°
Vertical FOV:		29.15°		26.99°

The diagonal fields of view are pretty much identical, but there are noticeable differences in the horizontal and vertical fields of view. In particular, this can make a difference when taking panoramas.

But the main gist of the article was about “equivalent aperture”. And somehow I'm not happy with the term. It's time for me to write my own page on the subject, which will (hopefully) grow.

So what's wrong with the DPreview page? The choice of aperture has two effects: it controls the amount of light entering the camera, and it modifies depth of field. Like so many others, this doesn't make a clear enough distinction between these two functions, and in my opinion it overemphasizes depth of field.

In particular: for exposure, there's already a clear definition of “equivalent aperture”: in more conventional terms, it would be called “relative aperture”, the diameter of the aperture as a fraction of the focal length. For example, a 25 mm aperture in a 100 mm lens is ¼ of the focal length, or f/4, where f represents the focal length. In other words, exactly what we used to have until people forgot what it meant and renamed it F4 (an “F-number”). No need for any further “equivalent”.

And then there's depth of field. That's a much more complicated matter. The first question is: how sharp is unsharp? That's defined as a circle of confusion, but what? In general it is chosen relative to the sensor size. My calculator currently uses a default circle of confusion of 8 μm, which in four-thirds measurements is about 1/2150 of the sensor width (17.3 mm).

The only parameters that enter into the depth-of-field equation are focal length, aperture, focus and circle of confusion. Keep them constant and increase the image size, say from four-thirds to “full frame”, and what happens? Nothing. But if you look at the circle of confusion as a fraction of the frame size, then things change. 1/2000 of 36 mm (width of a full frame) is 18 μm. And that makes a difference if you're looking at the entire frame. The comparison photos on page 3 show the big picture nicely. But that's not depth of field. It's extent of unsharpness, and it results from (effectively) viewing the full frame image from double the distance.

And there's the point. We don't just have exposure and depth of field. We have deliberate unsharpness as well, and that's not the same thing as depth of field. I need to think more about this, but one thing's clear: take, say, an Olympus E-1 (horizontal resolution 2,560 pixels, or 6.75 μm per pixel) and a Olympus OM-D E-M1 Mark II in high resolution mode (horizontal resolution 10,368 pixels, or 1.67 μm per pixel), and you have an argument for completely different depths of field, even though the sensor size is the same.