# Digital Camera Buyer’s Guide: Real Cameras

[Total: 1    Average: 5/5]

I’ve never owned a camera, but I want to get a ‘real’ camera (that doesn’t cost too much).
I’ve owned an entry-level camera already and I want to take the next step.

A: Usually called ‘bridge cameras’, ‘prosumer cameras’, or ‘micro cameras’, these cameras will allow you to have manual control over various parameters: the f-stop, shutter speed, ISO setting, etc. The lens is usually attached permanently. This type of camera can remain useful for about 3-5 years before technological improvements make it obsolete.  This is where life gets interesting- there is a huge range of options, performance, and pricing.  There is a bewildering array of options, and cameras can appear like a wallet, a beefed-up version of a point and shoot, or something that looks like a DSLR (digital single lens reflex). As a result, you should take some time to learn a few essential facts and concepts about optical imaging to determine what performance metrics are important to you. Furthermore, the amount of optional electronic processing available greatly increases- automatic face recognition, for example.  Whether or not these ‘features’ enhance or inhibit your ability to take a quality image depends entirely on you; so make sure you can at least return all settings to factory default or turn them off.  This section is longer than the other two, because it assumes you do not know any optics or imaging theory.

Focal length – this is one of three key concepts in imaging.  The focal length of a lens does not refer to the ability to focus.  The focal length is one of the design parameters of a lens, and relates to the angular magnification.  The angular magnification of a lens is equal to the ratio of back to front focal lengths. Thus, for example, a long focal length telephoto lens has a high angular magnification while a short focal length wide angle lens has a low angular magnification (sometimes less than 1). Angular magnification is not the same as the reproduction ratio: the magnification of the image with respect to the object. The reproduction ratio of most camera lenses is very small- the image of a person on the sensor is much smaller than the actual person, for example- even though the angular magnification may be very large (imaging someone who is far away).

By convention, the designated focal length of a camera lens can also be stated as ’35mm equivalent focal length’. This does not mean the actual focal length is useless information- it is needed to calculate the depth of focus and hyperfocal distance. If the actual focal length of the lens is specified and the sensor size is smaller than 35mm format, the field of view of the image will correspond to a 35mm image taken with a lens whose focal length is longer than the specified value.  For example, a 35-150mm zoom lens on a four-thirds camera will produce images that have the same field of view as a 35mm camera using a 70-300mm zoom- this can be calculated knowing the ‘crop factor’ of the sensor (discussed below). Knowing the 35mm equivalent focal length is helpful when composing your image- human vision is very nearly equivalent to a 50mm lens (35mm format).

One final note about focal length, regarding how your brain constructs depth information from a photo. The image is flat; your brain adds depth perspective to the image based on how your brain has learned to see with your eyes.  Images taken with a 50mm lens on a 35mm format will appear very natural to your brain.  Lenses with a shorter focal length will produce images that your brain will interpret as having exaggerated depth; conversely, telephoto lenses produce images that your brain will interpret as being compressed in depth.  Part of this is related to the way angular magnification changes the relative sizes of near and far, but it also depends on how your brain extracts 3-D information from a 2-D image.

f-stop/aperture setting: the second key concept.  The f-stop (and related concepts like numerical aperture) is probably the most important concept in imaging theory. Camera lenses can be set to a series of discrete values of the f-stop (the f-number).  The f-number sequence is logarithmic, and defined by first starting with ‘1’ (f = D) and then using aperture diameters that successively halve the amount of light passing through the lens:  the f-number sequence begins 1, 1.4, 2, 2.8, etc.  Photographic lenses are specified in terms of their focal length and the minimum f-number available (largest aperture diameter): for example a 50mm f/1.4 lens, or a 70-200mm f/5.6 zoom lens.  You don’t need to memorize the sequence, but it can be useful to know what ‘going up an f-stop’ means.  Some cameras allow for half-stop or third-stop increments, which is important for ‘exposure bracketing’ (see below).

Just as the focal length has nothing to do with the focus distance, the aperture diameter is not the size of the front element.  The aperture of a lens is a surface within the lens, and typically consists of an adjustable iris.  A primary effect of varying the f-stop is to vary the depth of field. Two other important optical results are that both lens aberrations and optical resolution increase as the aperture increases.  Most lenses on bridge cameras have moderately high f-numbers (say f/3.5 and up).  This, in conjunction with the small sensor size, keeps levels of aberration down and the depth of field high.  High f-numbers correspond to an accurate paraxial approximation (sin(q) = q) so the dominant aberrations are the ‘primary’ aberrations (discussed below). At high f-numbers, it is fairly easy to achieve excellent aberration correction, and images taken with these cameras will look good- uniform focus across the image, for example.  Two important exceptions are distortion and chromatic aberration- those do not depend on the aperture, and thus may be the primary residual aberrations.  Aberrations are discussed more fully below.

A note regarding zoom lenses: zooms have become the dominant consumer lens, but are significantly more complex than fixed focal length (‘prime’) lenses and difficult to discuss in an introductory essay. The way elements within a lens move during focus is much different than how they move during a zoom. Zoom works by changing the magnification of a front group of elements, a rear group, or both. If zoom changes only a front group, the change in the focal length of the lens is exactly the same as the change in the diameter of the entrance pupil, so the f-number remains constant during zoom. If the magnification of the rear group changes, the f-number will change as the lens zooms. In practice, most lenses do the majority of their zooming with the front group, allowing the zoom to retain most or all of the maximum aperture setting. Again, the lens specification should indicate by how much (if at all) the maximum aperture changes during zoom.

Exposure time– the third key concept.  On one hand, setting the exposure is a trivial matter- long enough to get sufficient light onto the sensor.  If you are imaging moving objects, it’s a little more complicated- a long exposure will result in motion blur (which may or may not be desirable), so if you want to freeze the motion (short exposure), you have to use alternate methods- a flash, increase the aperture, or increase the ISO setting.  The ISO setting is an adjustment to the electronic gain at the sensor- a higher ISO means more gain, which introduces more noise.  Some cameras let you operate in ‘aperture priority’ or ‘shutter priority’, which means you actively control one (f-stop or shutter speed), and the camera optimizes the image by adjusting the other parameters automatically.

Resolution:  Now we come to a concept that is terribly misunderstood and often the subject of spurious claims.  A detailed discussion about resolution is beyond the scope of this note, so for now we will simply distinguish between the maximum enlargement you can produce via a print and the ultimate resolution limit due to the lens itself.  The print size is different from the display size: 72 dpi (dots per inch) looks great on your monitor, but terrible on a print. The professional standard for printing is 300 dpi.  That may sound like a lot, but based on that specification, a 2 MP camera can produce a professional-quality 4 x 6 print.  An 8 MP camera has sufficient pixel count to produce a professional-quality 8″ x 10″ print.  Higher pixel-count (larger sized) sensors allow you to crop smaller regions (boring parts of the image) and still retain the ability to produce large professional-quality prints.

The ultimate resolution a lens and sensor can deliver depends on the f-number, the degree of aberration correction, and the pixel size.  For an aberration-free lens, points at the object are mapped to Airy disks at the image, the size of which are characterized by the distance from the central peak to the first minimum and is given by the Rayleigh criterion, which for visible light (0.5 micron wavelength) reduces to r = 0.6*(f-number) [in microns]. An f/4 lens produces an Airy disk radius of 2.4 microns, while stopping down to f/16 produces Airy disk radii of 9.6 microns. Pixel sizes should be not much larger than given by the Rayleigh criterion, or the sensor will limit the attainable resolution (Nyquist’s sampling theorem). For this example, pixel sizes greater than about 3 microns on a side will result in sensor-limited resolution at f/4.  A brief survey of pixel sizes currently in production indicate the typical camera has pixels 2-3 microns on a side, meaning the attainable resolution is more dependent on the lens than the sensor unless you are imaging with a fast (f/2.8 and below) lens.

Enlarging an image past the performance limits of the lens (or digital enlargement beyond the capabilities of the sensor) results in ’empty magnification’; blur circles simply become larger blur circles.  The rule of thumb is that empty magnification begins at 500/f-number. It is important to note that this rule of thumb is violated for camera sensors due to the Bayer filter and interpolation. Thus, using an f/4 lens on a 1/1.7″ sensor limits the ultimate size of the image to less than 2m on the diagonal; recall at 300 dpi, the maximum professional-quality print size will be about 8′ x 10′.  To be sure, you could make a large poster out of the image, but it will only look professional from a distance.

Crop Factor: The crop factor is the size of the sensor relative to the 35mm format. For example, a crop factor of 1.6x means the camera sensor diagonal length is 26.8 mm. All manufacturers use their own sensor format.  Thus, it is helpful to refer to the crop factor’ because the sensor is then expressed in terms of a standard- the 35mm standard.  To calculate the focal length for your camera, given the 35mm equivalent focal length, simply multiply by the crop factor.  Thus, a 200mm f/5.6 lens on a 1.6x sensor will actually appear as a 320mm f/5.6 lens mounted to a 35mm camera.  Note, the f-number has not changed- the focal length and f-number of a lens is an intrinsic property of the lens.  By changing the image size, different magnifications are needed to generate identical display sizes.

Depth of field– a precise depth of field calculation is difficult to perform, since perfect focus exists only in a single plane. This subject is covered in detail in many other photography sites, so we will not repeat them here. A (relatively) simple formula can be written down using lens and camera parameters: the actual focal length f, the f-number F, the magnification m, (the ratio of object distance to image distance), and the diameter of the ‘circle of confusion’. The circle of confusion is the size of the blur spot that your eye can barely resolve. Based on studies of visual acuity, c = 30 microns for 35mm format images. The formula is fairly straightforward:
$$DOF = \frac{2f\frac{m+1}{m}}{\frac{fm}{Fc}-\frac{Fc}{fm}}$$
where DOF is the distance range over which objects will appear in acceptable focus. For your camera, c = 30/(crop factor) microns.  As specific examples, using the 35mm standard, a 50mm f/1.4 lens focused 10 m away has a depth of field of 3.5m, while the same lens stopped down to f/11 has an infinite depth of field (near focus = 4.3 m) (this is discussed below, under ‘hyperfocal distance’).  By contrast, a 200mm f/5.6 focused 100m away has a DOF = 102m. A 1/1.7″ sensor using a 10mm f/5.6 lens will render all objects between 1.3m and infinity in focus. Small sensor digital cameras often do not have any out of focus components in an image. There are a multitude of free online DOF calculators available that have the relevant data for nearly all cameras on the market.

ISO: ISO stands for the “International Standards Organization”, and in imaging, refers to the sensitivity of film to light. The nomenclature carried over to digital imaging, and in this context refers to a level of gain (amplification) applied to the sensor output.  In conjunction with exposure speed, the ISO setting will adjust your camera’s light sensitivity.  Doubling the ISO setting doubles the sensitivity. This can be used to retain a fast shutter speed to capture a moving target. Again, increasing the ISO setting increases the amount of noise present, and different manufacturers use their own idea of how to reduce the noise levels. It’s not uncommon to find cameras with ISO settings up to 6400; ISO 102,400 is not unheard of.  Because digital technology is so different than film, this standard has come under renewed scrutiny; however as a rule of thumb daylight imaging should be performed at ISO 100-200, indoor at ISO 400-800, and nighttime imaging at ISO 1600 and above.

Exposure bracketing– getting a well-exposed image can be tricky, and there may not be a lot of time to adjust settings.  Photographers learned a long time ago to perform ‘exposure bracketing’.  Instead of taking a single image, they would take a series of images through a f-stop interval, for example 3 1/2-stop increments through a full stop (- 1/2, 0, + 1/2).  Digital cameras can achieve this by varying the f-number of the lens (shutter priority mode), the exposure time (aperture priority mode), or both. Some cameras allow for continuous shooting of an exposure bracket by simply repeated pressing of the shutter release.

Frame rate– if you are interested in photographing moving objects, you may be interested in how fast a camera can take continuous images.  The rates can vary, and some very clever autofocus routines have been devised to allow continuous autofocus while imaging a moving target.

Image Histogram– Now that you are becoming familiar with the parts of a camera and how to control the amount of light incident on the sensor, you should understand a basic ‘quality metric’ of the image- the histogram.  A histogram is nothing more than a graphical representation of the intensity levels in your image (a graphical representation of the dynamic range).  There may be individual histograms for each (r,g,b) color, or one overall histogram.  Either way, using the histogram will help ensure that your image is not underexposed (lots of black) or overexposed (lots of white).

Three ways to control the brightness of the image have been discussed- adjust the aperture, adjust the exposure time, and adjust electronic gain.  But, there are consequences to making adjustments to any of them- opening the aperture decreases the depth of field and increases the aberrations (discussed below), increasing the exposure time can lead to motion blur, and increasing the gain increases the amount of noise.  Learning how to control these elements in your image will enable you to take better photographs (if that’s a goal).

Post-processing– Often, these cameras will have a multitude of ‘on-chip’ image processing options available: you can adjust the contrast, sharpness, saturation, etc.  Whether or not you use these is up to you. Many manufacturers will include a basic image processing program bundled with the camera hardware.  In addition to commercial programs, a free open-source image processing program (ImageJ) is available that you can download and use to manipulate your images- correcting the brightness and contrast, or color balance, cropping, etc.

Pixel size vs signal to noise: here is another trade-off.  Smaller pixel sizes can increase the ultimate resolution and maximize the final print size, but smaller pixels also have smaller light-sensitive areas and thus the sensor needs more light to generate a good signal-to-noise ratio. As we saw above, pixel sizes smaller than 2 microns on a side will not generally increase the attainable resolution in digital cameras. Micro cameras found in cell phones have pixel sizes approaching 1 micron on a side, and intensified cameras used in low-light applications often have pixel sizes around 15 microns on a side.

Macro imaging: Macro imaging occupies the space between photography and microscopy.  Objects are small (but not too small), and like microscopy, images are usually described in terms of the reproduction ratio.  Macro lenses are designed to work close-focus and come in a variety of focal lengths. Longer-length macro lenses allow macro imaging of objects that are farther away: photographing insects at a distance, for example.  Because the lens operates at close focus, control of the depth-of-field is critical; often the aperture is set very small and a flash (or several flashes) are used to allow a reasonably fast shutter speed.

Autofocus: Autofocus is a complex multicomponent closed-loop control system consisting of a sensor (different than the image sensor), a control circuit, and a motor-driven lens.  The time needed to find best focus will vary with aperture setting.  Often, the manufacturer will have several autofocus modes: focus on the center of the image, find best focus over the entire image, or some other focusing algorithm. While some autofocus systems may be active (they emit IR light or ultrasound), most digital cameras use passive autofocus systems- either contrast measurement or phase detection. In general, phase detection is faster and more accurate, but both methods are constantly being improved. “Trap focus”, or “catch in focus” are mechanisms that allow the camera to use autofocus as a detector, acquiring an image when an object passes into the focal plane- this is very useful for photographing fast-moving objects.

Sharpness: another misused concept. At a minimum, “sharpness” implies a well-focused image using a well-corrected lens (aberrations are discussed below) with no motion blur. A related term is ‘acutance’, which describes the edge transitions between light and dark: a high acutance image can support very well-defined and rapid variations between bright and dark (high-contrast images).  Due to image processing in human vision, a high contrast slightly blurred image will be perceived as being sharper than a better focused, lower contrast image.

Memory cards: Most cameras require a memory card in order to store images. In addition to direct transfer of images from the camera to your computer, the camera will usually record the image onto a removable memory card that can permit much faster data transfer rates.  A typical standard format is Secure Digital (SD), and new formats are introduced with some regularity in an effort to increase data throughput.  The camera may come with a card, or it may not- in any case, it’s worth getting one with as much memory as you can afford.  In order to use the memory card with your computer, you will most likely need a ‘card reader’ that essentially converts your memory card into a USB memory stick.

Tags:
5 replies
1. phinds says:

question:

Crop Factor: The crop factor is the size of the sensor relative to the 35mm format. For example, a crop factor of 1.6x means the camera sensor diagonal length is 26.8 mm.

I've never heard of crop factor before but I don't follow that math at all. Can you expand on what this means / how it works?

Thanks

2. phinds says:

Diagonal of standard 35mm sensor (36mm x 24mm) is approx 43mm, so crop factor 1.6 means sensor diagonal about 43mm/1.6 which is about 27mm diagonal.

https://en.wikipedia.org/wiki/Crop_factor

Ah. Got it. Thank you. I do think that the assumption that this is immediately understandable is a bit of a stretch since the article (which is very good) is for beginners so @Andy Resnick, would a beginner be expected to know that 35mm => 35×24 ?

3. olivermsun says:

Ah. Got it. Thank you. I do think that the assumption that this is immediately understandable is a bit of a stretch since the article (which is very good) is for beginners so @Andy Resnick, would a beginner be expected to know that 35mm => 35×24 ?

36×24, and probably not, but it isn't so important to know these exact dimensions as long as you realize that's what people are using as the reference when they say a certain camera/lens has an equivalent zoom range of 28-200 mm or whatever. They mean the field of view covers the same range as a 28-200 mm lens would on a traditional 35 mm format camera.