How the sensor in your camera influences the images it produces…
The differences in size in commonly-used image sensors.
Full-frame: image sensor size 36 x 24 mm – which has the same surface area as a 35mm film frame.
APS-C: image sensor size between 21.5 x 14.4 mm and 23.7 x 15.7 mm in area.
M4/3: image sensor size 18.0 x 13.5 mm.
So why is the size of the sensor important? How does it affect image capture? And where does it have its greatest impacts?
Sensor sizes and aspect ratios
The first digital cameras adopted the 4:3 aspect ratio common to TV sets at the time. This aspect ratio is still used for the sensors in compact digicams, many smartphones, Micro Four Thirds system cameras and most medium format cameras.
Cameras that offer a range of aspect ratio settings obtain them by cropping the frame. So, before you can evaluate the ways the sensor influences imaging performance, it’s important to know the original aspect ratio of the sensor.
How a 4:3 frame is cropped to produce a video frame with a 16:9 aspect ratio. Pixels in the shaded areas above and below the video frame are discarded.
Once you know the aspect ratio of your camera’s sensor, the next step is to understand how it affects the lenses you use. The relevant term ‘crop factor’ relates the effective focal length of the lens to a familiar 35mm format.
Sensor sizes and crop factors
Most photographers recognise that the physical size of the sensor determines how much of the scene a particular lens can cover. The ‘crop factor’ comparing the sensor’s diagonal length to the diagonal of a 35mm frame is commonly used for describing the angle of view of lenses designed for 35mm sensors when used on cameras with smaller (or larger) sensors.
Camera and lens manufacturers commonly use crop factors to explain how images recorded by cameras with sensor smaller than a 35mm frame will appear enlarged or ‘zoomed in’ compared to when the lens is used with a 35mm camera. Thus, a 12-60mm lens on a M4/3 camera – which has a crop factor of 2x – will capture a quarter of the area of a 35mm frame and give the impression of producing a 2x enlargement with a zoom range equivalent to 24-120mm in 35mm format.
It’s easier to shoot close-ups with a shallow depth of field when your camera has a large sensor. This shot was taken with a 35mm camera with a focal length of 50mm and aperture of f/2. There’s no equivalent available for M4/3 cameras.
At the other end of the scale, suppose you were shooting landscapes with a 24mm wide-angle lens and wanted as great a depth of field as possible. You would be fine with a M4/3 camera using an aperture of f/16, although diffraction (see below) may begin to impact on image quality. However, with the 35mm camera, the equivalent aperture of f/32 would almost certainly be diffraction affected.
Sensor sizes and diffraction
Smaller sensors are more diffraction limited than larger sensors, primarily because the images from larger sensors don’t require as much enlargement to achieve the same print size. Any loss of sharpness resulting from diffraction is therefore less obvious.
If we take a fairly typical resolution of 20 megapixels, we obtain the following diffraction limits for six sensor sizes, ranging from 6x7cm medium format through 35mm, APS-C, M4/3 to 1-inch Type sensors.
- 6x7cm – f/26.3
- 35mm – f/12.3
- APS-C with 1.5x crop – f/8.1
- APS-C with 1.6x crop – f/7.7
- M4/3 – f/6.5
- 1-inch Type – f/4.8
These figures are directly related to the size of the pixels. Note that diffraction makes its presence felt gradually, which means half a stop on either side of these aperture settings is unlikely to look better or worse.
The lens itself may also be more or less susceptible to diffraction, depending on its ability to resolve fine detail. More expensive lenses tend to be better performers.
Pixel size, dynamic range and noise
Because they are able to capture more photons of light, larger photosites can generally also record a wider dynamic range with lower noise levels. Dynamic range describes the range of tones which a sensor can capture while retaining details in both highlights and shadows.
The recordable dynamic range depends on the number of photons received during an exposure and how that data is processed. Larger photosites will capture more light than smaller ones and the image processors used in cameras with larger sensors can generally extract more data.
The image on the left was captured with a M4/3 camera, while the one on the right was recorded with a smartphone with similar resolution but a much smaller sensor. Note the blown-out highlights (circled in red) and the blocked-up shadows (circled in yellow) in the smartphone image.
This is the reason photos taken in bright outdoor lighting with compact digicams and smartphones tend to have blown-out highlights and blocked-up shadows. As a rule, larger sensor with larger photosites will produce a higher signal to noise ratio ““ and a correspondingly wider dynamic range with lower image noise.
However, if two sensors that are the same size have different pixel counts, even if the images they produce look virtually identical when viewed at 100% on a screen, the sensor with the higher pixel count will produce a cleaner-looking print. This is because the degree of enlargement required for a given print size is less with a higher pixel count sensor so any associated noise has a higher frequency will appear finer grained.
In summary: even though they cost more and are heavier to carry around, cameras with larger sensors generally provide more control and greater shooting flexibility, particularly for shooting with a shallow depth of field. However, cameras with smaller sensors can be capable of achieving a comparable depth of field provided you remain within their limitations.
Article by Margaret Brown – see Margaret’s photography pocket guides