Making fine prints in your digital darkroom
Digital cameras
by Norman Koren

---

Site map/guide to tutorials Contact | News Making fine prints in your digital darkroom Understanding image sharpness and MTF Image galleries / How to purchase prints Photographic technique Image editing with Picture Window Pro A simplified zone system Digital vs. film

View image galleries

Search WWW Search www.normankoren.com

Related pages:

Digital vs. film | Tonal quality and dynamic range in digital cameras

Digital cameras are capable of excellent image quality; most digital SLRs have overtaken 35mm film cameras. Digital cameras vary enormously in size, shape, features, and operation. This page focuses on image quality; I can't keep up with all the latest models— there are simply too many. For detailed product information, news, and reviews, check the Links.

With digital cameras you don't have to buy film; you don't have to make one trip to drop it off for development and another to pick it up (or mail it out and wait). Of course you have to buy enough storage (cheap) and make sure to charge the batteries. I particularly like the ability to view images immediately after you've capture them. You can see if your composition is OK and you can check exposure with the help of histograms. For some reason, still a mystery, I find that I can make more consistently sharp images with digital than I could with 35mm film.

• Compact digital cameras. Sensors typically have between 2 and 8 megapixels. Sensor size is 11 mm diagonal or smaller (less than 1/16 the area of a 35mm frame). Most of these cameras can make excellent letter-size (8.5x11 inch or A4) prints; the best of them make decent prints up to 13x19 inches (Super A3). They come in a wide variety of shapes and sizes, though all are small compared to 35mm cameras. The smallest fit easily into a pocket or purse. The differ primarily in viewing systems: most have a separate optical viewfinder window; a few have electronic LCD viewfinders, which resemble SLR reflex viewfinders from the outside. Most have an LCD screen on the back that allows you to preview the digital image. But this feature comes at a price: most compact digitals have significant shutter lag— time delay between pressing the shutter and capturing the image, which can be as long as 1/2 second. Shutter lag is rarely mentioned in the brochures; it may make the camera unsuitable for action photography. It's getting better in the latest models. Many have close focusing (marco) capability. Compact digital cameras advanced rapidly until about 2003, when pixel count leveled off: image quality (exposure range and noise) tends to be compromised by making pixels smaller. Advances now consist of new features and incremental improvements in image quality.

Two compact digital cameras

Nikon 5700 A classic design employing a 180,000 pixel digital reflex viewfinder. 5 megapixels, 8x optical zoom: 8.9-71.2 mm f/2.8-4.2 (35-280 mm 35mm equivalent). Focuses to 0.8 inches in macro mode. Such designs typically have larger zoom ratios than cameras with separate viewfinder windows. Dimensions (WxHxD): 4.3x3.0x4.0 inches (108x76x102 mm); large for a compact digital. Weight: 16.9 oz (480g) without battery. Though its image quality is excellent, Michael Reichmann found its operation to be clunky.

Canon S45 / S50 A classic design employing an optical viewfinder window. 4 megapixels for the S45; 5 megapixels for the identical S50, 3x optical zoom: 7.1– 21.3 mm f/2.8-4.9 (35–105mm 35mm equivalent). Focuses to 3.9 inches (10 cm) in macro mode. Dimensions (WxHxD) 4.41x2.28x1.65 inches (112x58x42mm) excluding protrusions. Weight: 9.2 oz. (260g) without battery. The S50 would have about the same image quality as the Nikon 5700. Differences: its maximum focal length is shorter, it doesn't focus as close, but it's smaller and lighter. At low ISO speeds (50 or 100) these cameras have image quality approaching DSLRs for prints up to 8½x11 inches (A4 size).

• Digital single lens reflexes (DSLRs). Sensors typically have 5 or more megapixels with sensor size = 22 mm diagonal or larger. (The exception is the Sigma SD9/SD10, whose Foveon sensor is different from other digital cameras. It has either 3.3 or 10 megapixels, depending on how you count.) These cameras resemble 35mm SLRs and have interchangeable lenses— usually interchangeable with 35mm lenses of the same brand. They use the same types of shutters as 35mm SLRs, so you can't preview the image in the LCD screen. Shutter lag is typically much shorter than compact digitals and image quality tends to be higher because the larger pixels have lower noise. DSLRs fall into two categories:
  
  
• APS-C or Four-thirds sensor size.  5-12 megapixels. Sensor sizes from 22 to around 27 mm diagonal; half to 2/3 that of 35mm (43.3 mm diagonal). This means lenses designed for 35mm cameras have a smaller field of view; they have a "focal length multiplier" between 1.5 and 2. Although I originally viewed them as a stop on the way to full-frame DSLRs, I now realize they're here to stay. Their image quality equals or exceeds 35mm, and they're smaller, lighter, and much less expensive than the full-frame DSLRs (below). Several manufacturers now make lenses for this format: because they don't have to cover the entire 35mm frame they can be sharper, somewhat smaller and lighter, and (perhaps) slightly less expensive. And ultrawides are now available from Canon, Nikon, Olympus, Sigma, Tokina, and Tamron. My Canon 10-22 mm is equivalent to 16 mm in the 35mm format.
  
  
• Full frame (24x36 mm or close) sensor size:  Up to 16.7 megapixels. These cameras have resolution approaching or equalling medium format, and they're rapidly replacing medium format in professional applications. Because sensor manufacturing costs rise rapidly with size, they are expensive (around $5,000-8,000). Medium format sensors (near 645 size) are also available, but far more expensive. A number of cameras have been announced, but are not yet available as of April 2005.
  

An entry level digital SLR

Canon EOS-300D Digital Rebel Introduced in September 2003, the EOS-300D is a breakthrough for the price: a full featured 6.3 megapixel DSLR that takes all EOS lenses for under $1000 US, including a decent 3x zoom: 18-55 mm f/3.5-5.6 (29-88 mm 35mm equivalent). It physically resembles a small 35mm film SLR. Dimensions (WxHxD): 5.6x3.9x2.9 inches (142x99x72.4 mm) without lens. Weight: 19.7 oz (560g) without battery or lens. The basic 18-55 mm lens weighs 6.7 oz (190 gm); high quality lenses weigh more. Although you wouldn't expect it to equal a premium "L" lens, Peteris Treijs tested it using my chart and found that it was consistently sharper than Canon's consumer-grade 24-85 mm. The EOS-300D is a good choice for serious photographers on a budget— for ambitious beginners. Its size and weight would make it unappealing for most casual photographers. Compared to the compact digitals, it's faster (very little shutter lag), has better image quality, especially at high ISO speeds (it's still nearly grainless at ISO 400) and for prints larger than 8½x11 inches (A4 size), and can take the full range of EF lenses. My EOS-10D has similar image quality, but it has a few more features, greater durability (a metal frame body), and it weighs 10 oz (230 g) more: 27.9 oz (790 g) without battery or lens. The 300D isn't for everyone (no camera is). One of my readers found its flash settings to be lacking in flexibility: he swapped his 300D for a 10D. See Canon's description, dpreview.com's review, and Michael Reichmann's comments.

View image galleries How to purchase prints . . An excellent opportunity to collect high quality photographic prints and support this website . The image on the right was taken with the Canon EOS-10D. .

The key specifications that affect digital camera image quality

---

• Pixel count  This is the obvious one (5 Megapixels, wow!). The more megapixels the better, but watch out. Megapixels are not the only indicator of quality; pixel size matters. The table below relates pixel count to maximum print size for high quality images. There are more total pixels on the chip than active pixels. Rows and columns near the edges are not used for image capture. The number of active pixels is what counts; they're implied in the table below. 

How many pixels do you need for a high quality print?

Millions of pixels Print size (inches) Pixel count doesn't tell the whole story. Pixel size (which affects exposure range and noise) and lens quality also affect image quality. For sensors with small pixels (under 4 µm), resolution may be limited by lens quality. You can print somewhat larger for sensors with large pixels (over 6 µm, found in digital SLRs). Of course you always can print larger if you're willing to compromise. 1 2 3 5 6 11+ 4x5 5x7 8½x11 11.7x16.5 (A3) 13x19 16x24+

• Sensor size  Can be confusing because it is commonly specified by a number such as 1/1.8" or 4/3" that has little apparent relationship to sensor dimensions. Dpreview.com has a nice explanation. These numbers are based on the outside diameter of Vidicon tubes— vacuum devices used long before the advent of solid-state (CCD and CMOS) sensors. I'd like to see this old nomenclature replaced by the sensor's diagonal size— it's nothing but confusing. The diagonal dimensions of the sensor is roughly, but not exactly, 2/3 the specified size. Most small sensors have a 4:3 aspect ratio (the same as standard video monitors), but some are 16:9 (HDTV). 35mm (24x36mm) has a 2:3 aspect ratio.

• Pixel size  Normally specified as a linear dimension, closely related to pixel spacing or pitch. Not always easy to figure out from camera specs, but you can find it if you know the pixel count and sensor size. Listed in the table below for several models. Typical pixel spacings are 2.6 µm for inexpensive compact consumer cameras, 3.4 µm for compact "prosumer" cameras, and 6.8 to 10 µm for DSLRs. The active pixel area is the square of the pixel spacing times the fill factor — the fraction of the total pixel area covered by the active element. The fill factor is rarely specified outside of sensor data sheets. It has little effect on resolution, but is has some relationship to noise and sensitivity: the larger the fill factor, the better. Some sensors have microlens arrays to increase the effective fill factor by concentrating light in the center of the each pixel.

It turns out there is an optimum range of pixel spacing for high image quality.
 
• Small pixels suffer from increased noise (hence reduced signal-to-noise ratio), reduced exposure range (AKA dynamic range— fewer f-stops), and reduced sensitivity (lower ISO speed). These effects are most noticeable in pixels smaller than 4 µm. The exact relationship between noise and pixel size is difficult to quantify since there are several noise mechanisms, each of which scales differently.
 
• Large pixels suffer from aliasing— low spatial frequency artifacts that appear when the lens has significant resolution above the Nyquist frequency: 1/(2*pixel spacing). Aliasing typically manifests as Moiré patterns on images with high frequency repetitive patterns, such as window screens and fabrics. It can be reduced by anti-aliasing (low pass) filters, which unavoidably reduce resolution and increase cost. Most DSLR sensors, which have 6.8 µm or more spacing, have anti-aliasing filters (the Kodak DCS 14n is an exception). I believe they are absent in most compact digital cameras: with 3.4 µm or less pixel spacing, they aren't needed.
 
• Small sensors run into problems with lens diffraction, which limits total image resolution at small apertures. The f-stop where the image becomes diffraction-limited is proportional to format size (sensor diagonal). It's around f/16-f/22 for the 35mm format (43.3 mm diagonal). At large apertures— f/4 and above— resolution is limited by aberrations. There is a resolution "sweet spot" between the two limits, typically between f/5.6 and f/11 for good 35mm lenses. A 22 mm diagonal sensor becomes diffraction-limited around f/8 and an 11 mm diagonal sensor becomes diffraction-limited around f/4— the same aperture where it becomes aberration-limited. There is no "sweet spot;" the total image resolution at optimum aperture is considerably less than for larger formats.
 
• Large sensors cost more. No getting around it. That's the major reason inexpensive cameras use small sensors— popular 11 mm diagonal sensors have 1/16 the area of a 35mm frame. Of course cameras with small sensors can be made very compact.

You can find detailed sensor specifications in pages from Sony, Panasonic, and Kodak. If pixels size isn't available, you may be able to use the following equations.

pixel size in mm = (diagonal in mm) / sqrt(H² + V²), where H is the number of horizontal pixels and V is vertical pixels.
pixel size in microns = 1000 (diagonal in mm) / sqrt(H² + V²)

The optimum pixel size for high quality imaging seems to be in the 5-9 micron range found in DSLRs. Larger pixels have aliasing problems and can't take advantage of high quality lenses. Smaller pixels have more noise and less sensitivity, though they can still produce excellent images in compact cameras, particularly at low ISO speeds (50 or 100).

Digital SLRs will stick with 5-9 µm pixel size as they evolve towards larger sensors with more pixels. A 24x36 mm (44.3 mm diagonal) full-frame sensor with 6.8 µm pixel spacing (the same spacing as the Canon EOS-20D/30D) would have 18.7 megapixels— close to the holy grail for digital SLRs. The performance of such a sensor would approach medium format film (see the analysis in Digital cameras vs. film)— and indeed it does for the 16.7 megapixel Canon EOS-1Ds Mark 2— but it will remain expensive for some time. A full frame sensor with 6 µm pixel spacing (24 megapixels) is probably the most we'll see in the forseeable future. Reducing pixel size would increase both cost and sensor noise, and would only offer a resolution advantage for exceptionally fine lenses, which you can test for quality using Imatest.

Luminous-landscape.com has two excellent articles on pixel/sensor size: Counting megapixels by Michael Reichmann and Digital camera Image Quality by Miles Hecker.  

• Lens  Lenses are specified by their focal lengths and and maximum apertures (smallest f-stop). The following two definitions are for beginners.

• The focal length (normally specified in millimeters) indicates the lens's field of view. Short focal lengths tend to be wide angle; long focal lengths tend to be telephoto. Prime lenses have a single, fixed focal length. Zoom lenses have a variable focal length. The zoom ratio— the ratio of the longest to shortest focal length— is usually reported as a multiplier, for example, 4x. The majority of compact digital cameras have zoom lenses. The exceptions are extremely inexpensive or compact models (some very tiny). Most DSLR lenses were originally designed for film cameras. They come in a wide range of zooms and primes. Early zooms weren't as sharp as primes, but current premium zooms are outstanding. Consumer grade zooms can still be mediocre. Compared to zooms, primes may have be less expensive, have larger apertures, or have special attributes like macro (extremely close) focusing (with higher quality than zooms) or tilt/shift capability.
 
• The maximum aperture (smallest f-stop) of a lens is generally specified, for example, f/2.0, or f/2.8-f/4.5 for a zoom lens that would have f/2.8 at its shortest focal length and f/4.5 at its longest. Smaller f-stop numbers admit more light. (The f-stop is defined as focal length divided by the (circular) aperture opening.) The minimum aperture (largest f-stop) is rarely publicized. It is f/8 for most compact digital cameras and f/16 to f/22 for 35mm lenses. The inverse relationship between aperture and f-stop can be confusing for beginners.
.
The equivalent 35mm focal length is often specified for digital camera lenses. That's because many photographers are familiar with lenses for 35mm cameras: we know 28 mm is a fairly wide angle and 135 mm is a moderate telephoto. The actual focal length is the equivalent focal length times the sensor diagonal divided by 44.3 mm. For example, the Fujifilm FinePix F700 is specified as having 35-105 mm focal length (35mm equivalent). Since the sensor size is 1/1.7" = 7.6x5.7 mm = 9.5 mm diagonal, the actual focal length is (35-105 mm)*9.5/44.3 = 7.5-22.5 mm.

Specifications sometimes mention "digital" zoom. You can safely ignore this spec: it's equivalent to cropping and resizing in an image editor; it has nothing to do with a lens's capability. "Optical" zoom is all that really counts.

Unfortunately, lens quality is not mentioned in the specs. (Have you ever seen a lens described as "mediocre?" I assure you they exist.) You have to search out reviews or go by reputation. Or, with the Imatest program, you can test your own.

There is an optimum match between a lens and a sensor. If a lens is too good— if has strong MTF response beyond the sensor's Nyquist frequency— aliasing can be a problem and the lens's sharpness won't be useable. If the lens is not good enough, it won't be able to take advantage of the potential resolution of the sensor. The lenses and sensors in high quality compact digitals are well matched, but matching is hit and miss with DSLRs. In general, cheap consumer grade lenses aren't up to the capabilities of DSLR sensors. Premium lenses, such as Canon's L-series, match nicely.
.

Pixel size and diffraction Diffraction, a fundamental physical effect, limits a lens's performance at small apertures (large f-stop numbers). It is approximated by the equation, Rayleigh limit (lp/mm) = 1600/N, where N is the f-stop. At the Rayleigh limit, MTF = 9%; lenses have little response at higher spatial frequencies. At large apertures (small f-stop numbers) lenses are aberration-limited. Aberrations are a function of lens design— not a fundamental effect. Here we only consider diffraction. The highest spatial frequency a sensor can resolve is its Nyquist frequency, equal to 0.5/(pixel spacing). When a lens is stopped down so its Rayleigh limit is below the Nyquist frequency, the camera is limited by the lens rather than the sensor. For optimum quality (when extreme depth of field is not required), the aperture should be set at least one stop larger than the aperture where the Rayleigh limit equals the Nyquist frequency: NR=N = 3.2 * pixel spacing (µm). For a pixel spacing of 3.4 microns, typical of compact digital cameras, NR=N = 3.2 * 3.4 = f/11, so the aperture should be set at f/8 or larger. The corresponding aperture range for digital SLRs, which have pixel spacings between about 6.8 and 9 microns, is f/16-f/22. f-stop Rayleigh limit (lp/mm) Pixel spacing (µm) for Rayleigh limit = Nyquist 5.6 286 1.75 8 200 2.5 11 145 3.44 16 100 5 22 73 6.87 32 50 10 Depth of field at a given f-stop is inversely proportional to a sensor's diagonal dimension, so a compact digital camera with an 11 mm diagonal sensor has the same DOF at f/8 as a 35mm camera at f/32 (plenty).

• Timing lags  These include shutter lag and total (shutter+autofocus) lag. You can speed things up on many cameras by pressing the shutter part way to activate the autofocus. But shutter lag can still be considerable, especially on older compact digital cameras. It's much better in DSLRs. The 4 megapixel Canon EOS-1D SLR, reviewed December 2001, is the fastest: it feels like a film SLR. The Canon EOS-10D is almost as fast. Watch out: camera manufacturers aren't eager to pubilcize timing lags. You usually can't find it in the specifications, where it should be. It's frustrating to press the shutter, then wait... Even half a second can seem like an eternity when you're ready. Long enough to turn Cartier-Bresson into Derriere-Bresson, master of the post-decisive moment. If you feel modern now, you'll be post-modern by the time the shutter trips. Did I warn you about starting me on puns? No pain, no pun.

You can find autofocus and shutter lag for several camera models by probing deep into Dpreview.com's camera reviews. Look for pages titled, Timings & Sizes. They contain a lot of detail. Examples, shutter lag for the 4 megapixel Canon G3, reviewed December 2002, is about 0.1 seconds— fast. It is 0.2 to 0.4 seconds for my son's 3 megapixel Kodak DC4800, reviewed December 2000.
 
• Sensor A/D converter bit depth  The bit depth of sensor's Analogue-to-Digital (A-to-D) converter is usually larger then the 8-bit depth of standard image files. (16-bit depth is available for TIFF and a few other image file formats, but not for standard JPEGs.) Bit depth isn't always listed in specification sheets. When it's available, dpreview.com lists it as "A/D Converter" on the specification pages of its reviews. Bit depth is 12 for the Canon EOS 10D and 300D. It's 14 for the recently announced Sony DSC-F818, a premium 8 megapixel compact digital with an electronic finder.

Exposure (dynamic) range is related to both sensor bit depth and sensor noise. Sensor noise, as we've mentioned, is closely related to sensor size— the larger the sensor, the lower the noise. Increasing the bit depth increases exposure range only if sensor noise is low enough. In the case of the aforementioned Sony DSC-F818, its 2/3" (8.8x6.6 mm = 11 mm diagonal) sensor has a wimpy 2.7 µm pixel spacing— small enough to make me skeptical of the advantage of the 14 bit A-to-D converter. For comparison, digital SLRs have pixel spacings between 6.8 and 9 µm; a 14-bit converter would be more advantageous in a digital SLR.

To take full advantage of the potential exposure range of a digital camera, you should capture images in RAW format. for further explanation, see the (still unfinished) page, Tonal quality and dynamic range in digital cameras. Sensor noise and dynamic range can be measured by Imatest's Q-13 Stepchart module. 

How is ISO speed determined in digital cameras? The answer is found in Kodak Image Sensors - ISO Measurement (App note MTD/PS-0234), extracted from ISO standard 12232:1998. (Only 73 CHF— about $59 US— for 131 kB; such a deal!) There are two basic types of ISO measurement: saturation-based (also called "base") and noise-based. The saturation-based ISO is ISO = (15.4*f#2) / (L*t)    [Exposed so an 18% gray card has a level 18/106 of full scale. See below.] where  f# is the effective f-number of the camera lens, L is the luminance in cd/m2 of an 18% reflector (the familiar 18% gray card), and t is the length of the exposure in seconds. Saturation-based ISO corresponds to the camera ISO setting, which is controlled by the electronic gain of the system. You can use this equation to find the approximate luminance of an object, L =  (15.4*f#2) / (ISO*t), where f# and t are read from an 18% gray card. For a typical white surface (90% reflectance), L =  (3.08*f#2) / (ISO*t). Digital image sensors are linear devices: their pixel voltage (and the pixel level in a RAW image file) is proportional to the light energy reaching the pixel up to the point where it abruptly saturates. Now here's the heart of the matter. When an exposure is made according to the above setting, an 18% gray card has a voltage or pixel level 18/106 of full scale (e.g., a pixel level of 696 at the output of a 12-bit A-to-D converter, which can represent 4096 levels). In an 8-bit color space encoded for display at gamma = 2.2 (sRGB, Adobe RGB, etc.), the corresponding pixel level is 114 (from the formula, 255*(18/106)(1/2.2) ). Any portion of the image with luminance equivalent to more than 106% reflectiance (5.9x brighter than the gray card) would be saturated— pure white; burnt out... gonzo. Since a white card is about 90% reflectance, that doesn't leave much margin. The Kodak document says, "Very demanding applications might need more headroom for highlights, and so might use a higher number when calculating base ISO." That's why exposure compensation— equivalent to increased ISO speed— is often required to avoid burnt out highlights for "very demanding applications" such as ordinary sunlit outdoor scenes. The Kodak document has little to say about noise-based ISO measurement, which is related more to image quality than to camera ISO settings.

Other considerations

---

• Color quality  can be remarkably good with digital cameras, even compact models. I find the color quality to be outstanding, perhaps because the process of generating color with a digital sensor is so much simpler than with film. Colors are purer and saturation doesn't need to be enhanced to make the color satisfying (though the sticker on the parking meter and yellow curb reach 100% saturation). Digital camera color quality can be measured with the Colorcheck module of Imatest.

• Memory (storage)  doesn't affect image quality directly, but the more you have, the more high quality images you can store. If you don't have enough memory, you may be forced to store images at reduced quality or fiddle around deleting images before taking more. Virtually all digital cameras come with inadequate memory. You will need to buy more. There are two generic types: microdrives (tiny disk drives, typically 1 GB or larger; Hitachi dominates the market) and flash (nonvolatile semiconductor in several formats; get at least 256 MB for a compact digital; 512 MB for a digital SLR). Flash cards are the best choice in most instances— they're now available in up to 4 Gigabytes. You should be aware that not all flash memory is created equal. Check outRob Galbraith's CompactFlash Performance Data base before making a purchase. Some manufacturers rate speed relative to an audio CD: 150 kB/sec. 12x is 1.8 MB/sec, 24x is 3.6 MB/s, etc. Actual in-camera performance tends to be slower. Large amounts of memory is particularly helpful for storing images in RAW format— direct unprocessed data from the sensor chip. You can obtain the best image quality by processing RAW images offline— on a personal computer, which has much more processing power than a camera. With memory, the more the merrier!
  
• Additional storage.  My EOS-10D stored ~150 raw images on a 1 GB Compact Flash card, enough to meet my needs for a day of hiking— I can always review images and delete some if needs be. But it's not enough for an overnight trip. You'll need a laptop or a compact storage device. Laptops are the most versatile, but they're expensive, slow to boot up, power hungry, and too heavy to carry on a hike. Adorama lists several storage devices under Digital -  Memory -  Portable Hard Drives. The Nixvue Vista series comes with 20-60 GB and has a small color LCD screen that allows images to be reviewed and zoomed. It displays EXIF exposure information and histograms, and it supports Canon raw formats. $435-610, depending on storage capacity. The Delken PicturePAD series has similar capabilities. Adorama has its own, less capable but less expensive, Super DigiBin series (20-60 GB). One of my readers, Dennis Fisher, gives it a strong recommendation, though its battery life is somewhat short. I've heard particularly high praise for the SmartDisk FlashTrax devices, available with 20, 40, or 80 GB— slightly more expensive than Nixvue. ( I purchased a Dell laptop (model 5150) with a UXGA screen that has a larger viewing angle than most laptops (±70 degrees claimed), making it marginally adequate for image editing (most laptops are terrible), but I might still get one of the simpler compact storage devices for trips to remote regions.
  
• Dust.  Dust on the sensor can be a problem in digital SLRs with interchangeable lenses. The dust doesn't move out of the way when you advance the film. You must be careful when changing lenses, particularly in windy or dusty environments. For example, my friend Alan Ackoff, will never change lenses (even with film cameras) when you're photographing a (dry and dusty) New Mexico rodeo.  The first line of defense is a simple syringe— the plastic type with a large bulb. www.cleaningdigitalcameras.com is an excellent website with professional-level information on sensor cleaning.

Probably the easiest way of thinking of DOF is to relate it to 35mm cameras. Let N be the f-stop (aperture) on your digital camera, let N₃₅ be the f-stop on a 35mm camera that gives the same DOF, let d be the diagonal of your digital camera sensor (see the Sensor designation table, above), and let d₃₅ = 43.3mm be the diagonal of a 35mm image. Then,

N₃₅ = 43.3 N/d

On digital SLRs with a "multiplication factor" M, typically around 1.6x (Canon EOS D60, Nikon D100), DOF is larger than a 35mm image with the same field of view by a factor M.

Summary

---

• Size and weight. Larger is often better, though not necessarily. Many compact digitals can make outstanding 8½x11 inch prints and several can make decent 13x19 inch prints, though larger, heavier digital SLRs have the edge at that size. What are your ambitions? What are you willing to carry?
• Cost. Most of us are constrained by budget. What are you comfortable with? With a careful choice of equipment you should be able to make excellent 8½x11 inch prints, even on a student budget. If you're "middle class" you should be able to reach 13x19 inches without too much pain. But if you want to make larger very high quality prints, you'll have to spend some serious money. I can't quite justify the Canon EOS-1Ds: I'd need a 24 inch wide Epson 7600 printer to take full advantage of its quality.
• Lens. What is the zoom range and maximum aperture? What about quality? "Digital" or "combined" has little to do with the lens, and should be ignored.
• Viewfinder. You should handle cameras to see what you like. Viewfinders are similar in digital and film SLRs, but the digital SLR viewfinder image will be smaller if the sensor is smaller than full frame 35mm. I'm happy with the viewfinder in the EOS-10D. Electronic viewfinders in compact digitals are not to everyone's taste: they don't have the clarity of optical viewfinders, which they superficially resemble, but they're improving and they allow larger zoom ratios. Zoom ratios in cameras with optical viewfinder windows are limited to about 4x; 3x is typical. But these cameras are much more compact, and image quality can be extremely high.
• Other features. Spend some time studying reviews. Examples: The Minolta DiMAGE A1, a compact digital with electronic viewfinder and 7x zoom (the successor to the 7-series), has built-in image stabilization: a very nice feature, particularly if you don't like carrying a tripod. The Fujifilm FinePix F700 claims to have a better exposure (dynamic) range than other compact digitals. Dpreview.com disputes that claim, though its data shows a small advantage. There's always something new to learn.
• RAW capability. Not of interest to casual photographers, but important if you're serious. RAW images can yield better images quality in difficult lighting/exposure situations.
• Reviews. Search them out in the links, below. See the Sharpness comparisons page of the Imatest website.

Links

---

Imaging-Resource.com and Digitalcamerainfo.com both use Imatest (developed by yours truly) for measuring image quality in their reviews.
Steve's Digicams and Digital Photography Review and  also have extremely comprehensive reviews.
Luminous-landscape.com  Michael Reichmann only reviews a few cameras— he focuses on the cream of the crop. His reports are always stimulating.
Photo.net digital camera reviews
Digital Outback Photo
RobGalbraith.com  Excellent source of detailed technical news in areas such as support software and accessories. His CompactFash Performance Database is particularly interesting.
Ecoustics.com has a comprehensive list of reviews on other sites. Useful for researching.

Additional links for digging deeper:

DCReviews.com | megapixel.net (excellent French/English monthly) | Digital Camera Resource
John Tinsley (high-end professional digital cameras) | fredmiranda.com | Digital PhotoCorner
Nikondigital.org. Information and reviews, including Canon SLRs. From Moose Peterson and David Cardinal.
DigitalCamera-HQ.com  A nice site for searching and comparing models, with links to reviews and price comparisons.
Short courses in Digital Photography 
LetsGoDigital News in English, Dutch, and Spanish.
Digital Photography For What It's Worth (excellent articles on selecting and using digital cameras, geared towards beginners)

Images and text copyright (C) 2000-2013 by Norman Koren. Norman Koren lives in Boulder, Colorado, where he worked in developing magnetic recording technology for high capacity data storage systems until 2001. Since 2003 most of his time has been devoted to the development of Imatest. He has been involved with photography since 1964.