In earlier versions of this article I unashamedly confessed to being a “cheapskate Luddite” who was sticking to film because of the cost and hassles of switching to digital. Even so, I expressed a belief that digital cameras are the future of photography, and that they will ultimately relegate all film to the status that vinyl LP records and Kodachrome now have— specialized, niche products for die-hard traditionalists who prize the medium’s special qualities.

I officially retired my Luddite status in April 2005, with the purchase of a Canon Digital Rebel XT. But for several years before then I watched the market develop and periodically updated this brief survey of digital camera technology. I have now expanded it somewhat, to make it helpful for film users who might be thinking of taking the digital plunge as well as those contemplating the purchase of a first camera.

Note: The cameras, lenses, and prices I mention here were current at the beginning of April 2008. As with most digital technology, the digital camera market is very “dynamic.” It’s all but impossible for anyone who isn’t a full-time reviewer to keep track of it, let alone maintain a comprehensive and up-to-date survey. I can only update this section sporadically, so the specifics of prices and models are almost certain to have changed by the time you read this. But the discussion does reflect the current state of technology, and the relative differences between the prices are likely to remain useful for comparison.

Digital Camera Taxonomy

I classify digital cameras into three distinct species. First, there’s the (relatively) inexpensive compact point-and-shoot digicam. These often lack the traditional squinty viewfinder— you compose and view your pictures on a smallish color liquid-crystal display (LCD) on the back of the camera, which provides a continuous “live” display of what’s in front of the lens as long as the camera is turned on. A digicam is an ideal choice for the family or vacation snapshooter, the 21st century’s answer to the 20th century’s box Brownie. But the better digicams are good enough for serious and professional photographers who need an unobtrusive supplement to their DSLRs.

There are even disposable digital cameras that store a limited number of images internally. When they’re returned for “processing,” you get a CD of JPEG files along with prints. An increasing number of digicams are built into cellphones. Most have lower resolution and image quality than stand-alone cameras, but that’s gradually improving. The proliferation and convenience of cellphone cameras may eventually render compact digicam obsolete.

The second species is the digital SLR (DSLR). Conceptually, this is a 35mm single lens reflex camera body adapted to use an electronic sensor instead of film. It accepts the same lenses as the manufacturer’s film SLR “system.” (Olympus and Panasonic offer the “Four Thirds” format that owes no allegiance to any “legacy” camera system.) DSLRs have all but supplanted their film-based ancestors among professional photographers, who prize the flexibility and efficiency combined with image quality that can be better than film. The prices of these cameras are continuing their inevitable descent, but an “inexpensive” digital SLR body still costs between $400 and $1,200, not including lenses, memory cards, or external storage.

For a while it seemed that a third species, the middle-ground “electronic viewfinder” (EVF) camera— sometimes called a “SLR-like compact” or a “bridge camera”— would provide a compact and affordable alternative to the DSLR. These cameras have zoom lenses with a wide range, typically equivalent to 24-140mm or 36-432mm on a film SLR. They offer all the exposure control of an SLR, along with typical digicam features such as movie clips and sound recording. The small size and wide range of their lenses makes the cameras much smaller and lighter than DSLRs, which would require two or more lenses to get a comparable range of focal lengths. The fixed lenses offer the additional advantage of reducing the risk of getting dust on the sensor, an inevitable problem with interchangeable-lens DSLRs that can be minimally annoying or show-stopping.

But improvements to both DSLRs and digicams are now conspiring to make the EVF camera irrelevant. Entry-level DSLRs are becoming as small and light as EVF cameras, at close to the same price when sold as a package with a cheap but entirely adequate “kit lens.” They also offer direct optical viewing through the lens, which is much more pleasant than squinting at the coarse LCD inside the “electronic viewfinder.” And digicams now feature zoom lenses with nearly the same wide range as EVF cameras, along with enhanced resolution.

Tiny Sensors, Image Stabilization, and Flash

EVF cameras and digicams share a significant drawback. They use very small sensors, about the size of a fingernail or a frame of old-fashioned Super 8 movie film. The little sensors allow manufacturers to design compact cameras with small and light lenses. But the unavoidable physics of packing many megapixels into a small space makes images much “noisier” than DSLRs with larger sensors. DSLRs can produce images without objectionable noise at the equivalent of ISO 400 or even 800. Current EVF cameras and digicams are effectively limited to ISO 64 or 100. At higher settings they produce either colorful flecks of noise or blurring from the camera’s attempt to eliminate those flecks. Manufacturers seem compelled to include continually-escalating but questionably useful ISO settings purely for marketing reasons.

Barring a revolutionary improvement in sensor technology, image stabilization provides the most effective way around the speed limitation of tiny sensors. It uses a miniature accelerometer to track the camera’s motion and provide the input to a computer that moves the sensor or lens elements in the opposite direction. By compensating for camera movement it allows shutter speeds up to about three stops slower, effectively increasing the camera’s usable sensitivity at low ISO settings and avoiding the noise or blurring of a higher ISO. Image stabilization (also called “vibration reduction,” “optical stabilization,” “shake reduction,” and possibly other names) is an essential feature to look for in a digicam or EVF camera, and many models include it.

But be sure the camera you’re considering actually has a real image stabilization system. Some digicams claim to have “digital image stabilization” that is actually nothing more than a high ISO setting. In a few digicams, the “digital image stabilization” is special processing in the camera’s firmware that attempts to reduce the appearance of motion blur when it creates JPEG files. I have found software motion blur reduction less than adequate with either Photoshop CS3’s Smart Sharpen tool or the specialized Focus Magic plug-in. Both tools are processor-intensive, requiring half a minute or more to run on a decently fast desktop computer. So I’m very skeptical about the capability of any motion blur reduction algorithm that can be implemented with a compact camera’s rather modest computing resources.

Be aware that image stabilization is neither a perfect nor a complete solution, since it only compensates for shaky hands and not subject motion. Capturing a late-afternoon soccer game or an indoor ballet recital will still require a high ISO setting. That gives you a higher shutter speed, which is the only way to get a sharp picture of a fast-moving subject so you’ll capture your kid scoring that goal instead of a smeary blur. The pipsqueak built-in flash won’t help, since its useful range extends no more than three or four meters.

But that flash can still be very useful. On a bright sunny day, when you’re taking snapshots of your family in front of the Casa Grande at Hearst Castle or an erupting geyser in Yellowstone, always turn on the flash before you tell them to say “cheese.” By “filling in” the harsh lighting, the flash will make each face distinct and visible. You’ll have a better chance of getting a cherished memory instead of a bright scene with some dark unrecognizable people in front of it. On an overcast day, the flash can prevent the white sky from tricking the camera’s meter into turning your kids into silhouettes. Whatever kind of camera you’re using, turning on the flash is about the simplest thing you do to improve your family snapshots.

DSLR Sub-Species

Manufacturers of digital SLRs have created a bifurcated, strictly segmented market. The most expensive professional-level cameras have “full-frame” 24x36mm sensors, the same size as a 35mm film frame. They produce 12 to 21 megapixel files, and cost between $2,200 and $8,000. Canon’s 12.8 megapixel 5D, hailed as a breakthrough in “affordability,” still costs around $2,200. With these cameras, lenses work exactly as they would with film.

The more affordable 6 to 12 megapixel amateur-oriented cameras have a smaller sensor, about 22x15mm. These sensors are often called “APS-C” because they’re about the size of a frame of nearly-obsolete APS film. An APS-C sensor uses only the center portion of the “image circle” a standard-format lens produces. The resulting cropping “magnifies” the effective focal length by a factor between 1.5 and 1.7.

The cropping can provide a real advantage for telephoto users, since a 200mm lens becomes the equivalent of 320mm. A 50mm “normal” lens that has a field of view similar to the eye’s on a 35mm or full-frame camera becomes an 80mm portrait lens. But it’s an equal and opposite disadvantage if you like wide-angle lenses. A 28-135 wide-to-moderate zoom turns into a normal-to-telephoto 45-210mm; you’d need a 17mm actual focal length to get the 28mm equivalent view. A 22mm lens would have an effective focal length of 35mm; you’d need a 14mm lens to get the equivalent of 22mm. Because good lenses that short need many elements to correct distortions, they tend to be bulky, heavy, and expensive.

Canon and Nikon have the largest share of the DSLR market. Canon’s APS-C DSLRs include the 10-megapixel 40D ($1,140) and Digital Rebel XTi, called the 400D outside North America ($520). Some time in the spring of 2008, the 12-megapixel Digital Rebel XSi (called the 450D outside North America) will displace the XTi as leader of the (Digital) Rebel pack. Meanwhile, the 2005-vintage 8-megapixel Digital Rebel XT (called the 350D outside North America) is still available and an excellent bargain at $395. Nikon’s competitors are the 12-megapixel D300 ($1,800), and the 10-megapixel D80 ($730) and D40x ($535). The older 6.1-megapixel D40 is Nikon’s bargain entry at $500, which includes an 18-55mm “kit” lens.

Sony and Pentax each have a range of DSLRs with resolutions and prices comparable to Nikon and Canon models. But the DSLRs from both manufacturers have image stabilization built in to the sensor assembly, so it works with any lens. That gives them an advantage over Canon and Nikon’s offerings, which provide image stabilization only when you use a lens with that feature.

Which one of these “camera systems” is “the best”? Unfortunately, I can’t give you an easy (or objective) answer to that question. It all depends on what you consider important. For example, if you already have an investment in lenses from a “legacy” system, it would make sense to start by considering that manufacturer’s cameras— provided they’re compatible with your lenses. Sometimes a special feature is the deciding factor. One reason I chose the Canon EOS system in 1989 was the unique “Depth” mode that automatically sets an aperture and focus distance that keeps everything in focus between the near and far focus points you select (I discuss this in greater depth here). But Canon seems intent on deprecating this valuable feature. The APS-C DSLRs all provide a much less useful “automatic” version of it, and none of the full-frame DSLRs (including the “affordable” 5D) include it at all. If I were starting from scratch today, I would be inclined to consider Sony or Pentax because of the image stabilization built into the camera body. But the tradeoff is a smaller selection of lenses— from the manufacturers and from third parties— than is available for Canon or Nikon cameras.

With one exception, DSLRs produce images in the traditional 3:2 aspect ratio (width and height) of 35mm film. Olympus Evolt DSLRs use the Four Thirds format. Developed by Olympus and Kodak, it’s a “legacy-free” system designed from the outset for digital cameras. The name comes from its distinguishing characteristic, a sensor smaller than APS-C with a squarer 4:3 aspect ratio and a cropping factor of 2. A smaller, squarer sensor needs a smaller image circle than APS-C, which theoretically means more compact and lighter lenses.

Olympus and Kodak apparently intended Four Thirds as a universal standard that the industry would embrace (and license), thereby solving the decades-old problem of incompatible SLR lens mounts. But the industry seems to have greeted it with a collective yawn. Although six companies besides Olympus are officially part of the “Four Thirds consortium,” only Panasonic has so far adopted it for their Lumix DSLR (there is also a Leica model based on the Panasonic camera). The format’s touted advantages haven’t panned out in practice. Four Thirds lenses aren’t significantly smaller, lighter, or cheaper than those for “legacy” formats; and the smaller sensor theoretically invites more noise, although that apparently hasn’t been a significant problem.

Four Thirds DSLRs share their 4:3 aspect ratio with digicams and EVF cameras. It exactly fits standard analog television screens and CRT monitors, but not the 16:9 digital HDTV sets or 5:4 and 16:10 LCD monitors that are rapidly replacing them. So much for “legacy-free”— unless you mean the standard “legacy” print sizes that it doesn’t fit either (but millions of digicam users also have that problem). The available Four Thirds cameras don’t offer anything that distinguishes them from Nikon’s or Canon’s more conventional ones. So I really don’t see much advantage to the format. It’s just the latest of many mutually-incompatible camera “systems,” with a “digital” orientation that seems more valuable to marketeers than to users.

Lenses Galore

Manufacturers are addressing the problem of wide-angle lenses for APS-C sensors with special “digital” lenses that have a reduced image circle. If you put them on a full-frame DSLR or a film SLR you’ll get very severe vignetting— a circular “periscope” image with black corners. In theory these lenses should be smaller and lighter than their full-frame equivalents, but in practice it hasn’t worked out that way.

At the low end is Canon’s 18-55mm f/3.5-5.6 zoom. It was first available only as a kit with the extinct original Digital Rebel (called the 300D or Kiss Digital outside North America); it’s often sold in a kit with the newer Digital Rebel XTi/400D. Small, light, slow, and plasticky, it more than slightly resembles the discontinued 22-55 f/4-5.6 full-frame zoom intended as the “kit lens” for an extinct APS SLR. It provides the equivalent of 29-90mm. The “micro-USM” motor version of the 18-55 is available for $140. A newer version that includes image stabilization costs $175. Depending on who you want to believe, the image quality of these lenses is either respectable or horrible. Like all of Canon’s reduced-circle lenses, these “kit” lenses have a special “EF-S” mount that fits only on newer APS-C DSLRs (which can also use full-frame lenses in the regular Canon “EF” mount). Canon’s high-end “normal zoom” is a 17-85mm f/4-5.6 EF-S with image stabilization that costs $515.

Nikon’s latest 18-55mm f/3.5-5.6 kit lens with “vibration reduction” sells for $200. An older version without image stabilization costs $120.

Sigma has a similar reduced-circle 18-50mm f/3.5-5.6 zoom for $109 that fits Canon, Nikon, and Sigma DSLRs. Its companion is a lightweight 55-200mm f/4-5.6 telephoto zoom for $150. There’s also a larger 18-50mm f/2.8 “professional” version for $419. Sigma and Tamron both have 18-200mm f/3.5-6.3 lenses for $380 and $399, respectively. The Sigma seems to get better reviews than the Tamron; tests in Popular Photography magazine show that both perform poorly at 200mm but quite well at shorter settings, with the Tamron having slightly more distortion. Sigma has a newer “OS” version of this lens with “optical stabilization” that’s supposed to offer improved optical performance as well; it costs $500. For the same price, Tamron’s 18-250mm f/3.5-6.3 Di-II LD Aspherical IF Macro forgoes image stabilization but has an extra 50mm at the long end.

I had originally planned to get the (discontinued) Sigma 18-125 with my Rebel XT, but I could not find any store in the Los Angeles area that stocked it. The stories I’ve read from owners of this lens suggest that Sigma has quality control problems. A good “copy” is very good, but too often you don’t get a good copy. That has made me reluctant to buy it from one of the mail-order vendors that do stock it.

At the high end of ultra-wide reduced-circle lenses is Nikon’s expensive ($920) 12-24mm f/4 zoom, for Nikon and Fuji DSLRs. Canon’s equivalent is a 10-22mm f/3.5-4.5 EF-S zoom for $690. Tokina, Tamron, and Sigma have some more affordable wide reduced-circle lenses: Tokina’s 12-24mm f/4 zoom has received good reviews and sells for $500. Sigma’s 10-20mm f/4-5.6 sells for $500. Tamron’s 11-18mm f/4.5-5.6 costs $569. I chose, and can recommend, the Tokina (read more about it).

Sigma’s 12-24mm f/4.5-5.6 ultra-wide zoom ($689) covers a full 35mm frame. It has the drawback of a very large front element and a permanently attached hood that doesn’t allow the use of a polarizing filter. Although the lens does well in magazine tests, and many users are happy with it, I have read a number of complaints about quality control problems. If you buy one, get it from a store that makes exchanges easy if you aren’t satisfied with your “copy.”

Memory and Storage

Digital image storage is one problem that technology is only slowly solving. For the sort of travel photography I do, it’s a simple matter to carry around, say, 20 rolls of 36-exposure film. That fits 720 high-resolution images into a small space that requires no additional equipment besides the camera.

Digicams avoid this problem because they compress their images into manageable JPEG files. An entire vacation’s worth of snapshots could fit comfortably on a tiny memory card. But for the best quality with a DSLR or EVF camera (and a few high-end digicams) you’ll want to use “raw” format that stores data directly from the camera sensor (see Raw or JPEG? for the reasons why).

A “raw” image from an 8 megapixel camera requires between 8 and 10 megabytes of storage. So the digital equivalent of those 20 rolls of film could require a bit more than 7 gigabytes. An 8 gigabyte Compact Flash card costs between $100 and $180, depending on its rated speed. 4 gigabyte Compact Flash card costs between $40 and $90. Digicams and a few DSLRs use the smaller Secure Digital (SD) cards. Their cost is similar to Compact Flash. The largest-capacity memory card currently available is a Transcend 32-gigabyte Compact Flash card that costs around $175; but cameras introduced before 2007 may not be able to use its full capacity.

When comparing memory cards, be aware that a more expensive card with a higher rated speed may not necessarily provide better performance with your camera. Most Canon DSLRs, for example, do not support the “UDMA write acceleration” necessary to take full advantage of the fastest cards. And even if your camera can write data at an expensive card’s full rated speed, it may not matter unless you regularly take “bursts” of fast-action pictures (e.g., sports, birds, auto racing). Rob Galbraith has a newly-updated CF/SD Performance Database that evaluates the performance of various memory cards with 15 current DSLRs. In general, if a faster card costs only slightly more than a slower one, you probably can’t go wrong with the faster card. But if the price difference is substantial, contact the camera manufacturer to confirm that the camera can actually write data at the higher speed. And buy major-brand memory cards (e.g., SanDisk, Lexar) only from reputable vendors. There’s a thriving trade in counterfeit memory cards on eBay and on other private-seller sites. If the price seems too good to be true, it almost certainly is.

When I got my Rebel XT in 2005, Compact Flash cards with sufficient capacity for a one-week trip would have cost $900. Although the prices of memory cards have dropped dramatically since then, it can be impractical for many photographers to buy and carry enough memory cards for an entire trip’s collection of raw files. Professional photographers often bring laptop computers; that lets them not only store their pictures but edit and e-mail them to impatient customers. If you don’t have a laptop computer, or if you don’t want to carry one on vacation, there are several models of basic portable 80-gigabyte hard drives that cost about the same as a 8-gigabyte memory card. They include a built-in memory card reader and the firmware necessary to transfer the card’s contents. At the end of the day, or perhaps during a lunch break, you can insert a memory card from the camera, upload the contents, and erase the memory card for reuse. Some models can do a byte-by-byte comparison of each file on the card and the disk to verify the integrity of the upload. More elaborate (and expensive) devices include a color LCD for reviewing images.

If you’re the worrying type, a stand-alone portable DVD burner that can write disks directly from a memory card might be more appealing than a portable hard disk. Several such devices are available in the $200-300 range. Carrying a bunch of disks is less convenient than storing images on one small hard drive. But it’s probably the safest available storage option, particularly if the device verifies disks after writing them. Like film, a DVD only needs to be kept away from heat and direct sunlight. Unlike film, DVDs aren’t affected by “security” x-rays (although the thumbprint of a zealous airport screener might cause problems). The difficulty is that the physical size of the disk makes the burner bulky and heavy. If you’re flying with just a carry-on bag, or if you’re otherwise traveling light, one of these devices plus the disks can take up a lot of scarce space and weight.

I ended up choosing a portable hard drive, the (discontinued) 40-gigabyte SmartDisk FlashTrax, rather than a DVD burner. The size, weight, and bulk were the deciding considerations. The FlashTrax is about the size of a small paperback book and weighs 340 grams. That’s compact and light enough to fit in my rather small camera bag, and it includes a color screen to display pictures (along with a remote control and cable to display the pictures on any television set with a video input). The FlashTrax has since been replaced with the slightly larger FlashTrax XT. And I have since taken advantage of the much-reduced cost of memory cards and retired my FlashTrax.

Raw or JPEG?

Many digital camera users find the concept of “raw” files mystifying. Why is it worth using up to 10 megabytes (for a typical 8-megapixel image) for a raw file that can only be read with special software, while a JPEG file of the same resolution uses a quarter of that space and can be read by just about any imaging software?

To clear up this mystery, you need to know how a camera creates a JPEG file. Once the shutter closes, the camera gathers the 12-bit or 14-bit data from all the “photo sites” on the sensor and stores it internally. Then it converts this sensor data into values of red, green, and blue for each of the millions of pixels in the picture. That’s a rather complex process because the sensor actually “sees” in black and white. The individual “photo sites” are covered with either red, green, or blue filters and arranged in a fixed mosaic pattern. The camera applies the white balance, contrast, sharpening, saturation, color tone, and other parameters you’ve set (or their defaults) to the image, reduces each pixel to 8 bits per color, and applies “lossy” compression. Lossy compression significantly reduces the file size by taking advantage of the limitations of human vision, discarding parts of the image your eyes and brain won’t miss. Finally, it writes the compressed file to the memory card and purges the original sensor data.

Conversely, when writing a raw file the camera takes the same data from the sensor, applies lossless compression— when you uncompress the data you get back exactly what you started with— and writes that directly to the memory card without any further processing. The raw file also records the parameters you set, and might include a small JPEG version of the image to allow easy previews.

Notice that to get the smaller file size, the process of creating a JPEG file permanently discards much of the data the sensor captured. Still, if you got the exposure, contrast, and color balance right when you took the picture, the resulting image will display and print very nicely. But what if you didn’t get those parameters right? If you try to make significant adjustments to the exposure, contrast, color balance, or other parameters in your image editing software, the absence of the data the JPEG process discarded can cause visible defects like cartoonish color (“posterization”), bands of color, blurring, or grain.

A raw file contains all the data the sensor captured. Appropriate software (a raw converter) can translate it into colored pixels much as the camera does. But it can create a TIFF file with the full 12-bit or 14-bit color of each pixel. JPEG must reduce each pixel to eight bits. So you have much more data available to adjust the image without creating artifacts, and particularly to correct exposure that might not have been perfect. You also need not worry about setting the right white balance in the camera, since you can adjust that on your computer at your leisure. (See A Bestiary of File Formats for a discussion of TIFF, JPEG and other common types of files.)

A raw file is something like an undeveloped film negative that the raw conversion software “develops” to create an image file (usually in TIFF format). You can then control this “development” process— and do it as many different ways as you want— to make a final “negative” that corrects any flaws, looks the way you want it, and is ready for “printing” in an image editor. You can also take advantage of possible improvements in raw conversion technology by simply “re-developing” your raw files with new software. By that analogy, a JPEG file is a Polaroid^® print that you pretty much have to accept just the way it pops out of the camera.

Does the larger size, inconvenience, and the need for special software really justify using raw files? That depends on what you need from your pictures. Cameras often do a very good job of producing JPEG files. When I’ve tried shooting both raw and the highest-quality JPEG files with my Canon Digital Rebel XT, I honestly couldn’t tell the difference between the JPEG and the raw file converted with parameters “as shot.” The compression artifacts and reduced detail only became visible when I magnified the JPEG enough to make the image break up into visible pixels.

If you just want small prints for a scrapbook or pictures of the kids for e-mailing to Grandma, raw isn’t worth the extra storage space and hassle. But if you want the full image quality your camera is capable of producing, along with the most control over what the final image looks like, raw is definitely worthwhile. A useful approach might be to use JPEG to maximize storage space for ordinary snapshots. Then switch to raw for detailed scenic shots that are likely candidates for enlargement, or when odd or mixed lighting (light bulbs or candles) is likely to require adjustment in post-processing.

About Raw Converters

As I’ve mentioned, you need special software— a raw converter— to “develop” the sensor data in raw files into pictures. Although any camera capable of producing raw files will almost always include the manufacturer’s raw converter in the box, you’ll probably want one or more third-party raw converters. Camera manufacturers are usually quite good at their specialty of designing cameras, scanners, and other hardware. But their software to support that hardware is too often lacking, as it’s not their “core competency.” Other software developers can often do better.

I’ll use my Canon Digital Rebel XT as an example. The camera comes bundled with two separate programs that can convert its raw files. ZoomBrowser EX is a friendly “one-stop-shopping” package for snapshooters. It can view and sort image files, do simple post-processing, print, e-mail, write files to CD, and save images for the Web or desktop “wallpaper.” It processes raw files with a function called “Raw Image Task” (RIT). RIT emulates the camera’s JPEG processing, although it can also produce 8-bit or 16-bit TIFF files. It provides the same controls as the camera, with sliders for exposure compensation and color balance, and presets for white balance, contrast, saturation, sharpness, and “color tone.” The lack of real control beyond the presets negates much of the advantage of using raw files. Saving the output to a TIFF file yields images that look identical to the camera’s JPEG files— which means they look very good— but without the lost data or compression artifacts. One thing to note is that RIT modifies the raw files it processes. I much prefer leaving the original files alone, and putting setting information in a separate file. Nikon’s equivalent “snapshooter” software is called Nikon View NX (which replaced Picture Project in 2008).

Digital Photo Professional (DPP) provides more control over raw file processing, including levels and curves adjustments. But it was written by a completely different team and uses completely different algorithms to process the sensor data. Inexplicably for software intended for professional or “serious” users, the version that ships with the 20D, Rebel XT and earlier DSLRs produces color moiré and artifacts with small details, and turns high-contrast black lines green. It’s quite possible that these deficiencies are visible only under pixel-peeping test conditions and high magnification, but cause few problems with real-world pictures. The current version 3 of DPP, which Canon introduced with the 30D, seems to have fixed these problems; but its user interface remains clunky and inconvenient. It’s available as a free download for users of earlier cameras. If you have a Nikon DSLR and you want the manufacturer’s professional-grade raw converter, you’ll have to pay an extra $180 for Capture NX 2.

Adobe’s Photoshop, Lightroom, and Bridge are fully integrated with the excellent Adobe Camera Raw (ACR). It converts raw files from over 150 different camera models, and Adobe is frequently adding new ones. Photoshop Elements also uses ACR, but with a simplified interface that doesn’t include certain advanced features. If you use any of those Adobe products, you’ll probably find that ACR fits more conveniently into your work flow than the manufacturer’s raw converter. Corel’s Paint Shop Pro also has a built-in raw converter, but it’s one of that program’s weak links. The way it’s integrated into the image editor doesn’t allow full control over the processing of raw files, and Corel seems incapable of providing timely updates for new cameras between their annual major releases. Some other third-party raw converters are Bibble, the quite expensive Capture One Pro, and the free Raw Therapee. Open-source dcraw is intended for Linux command-line wizards but available for Windows and various other platforms; it’s also the basis of numerous other raw converters.

Adobe Camera Raw 4.4.1 produced noticeable red and green stripes when rendering this raw file. This cropped section of the image is 75% of full size. At normal magnification, the stripes give edges an unnatural jagged appearance. ACR 4.5, released the week after I first published this discussion, produces the same artifacts.

Canon Digital Photo Professional 3.4 rendered the raw file with very faint striping, visible only at high magnification.

Whichever raw converter you prefer, I recommend collecting as many raw converters as you have space (or money) for. The reason is that demosaicing— the process of converting sensor data into images— is a complex undertaking. Developing the necessary algorithms is often as much an art as a science. Every raw converter takes a slightly different approach to its demosaicing algorithms. The art and science of raw conversion has matured to the point where you should get very similar output no matter which software you use. But the differences are enough so that you will inevitably run into some images that will yield better results with a raw converter other than your preferred one.

I use Adobe Camera Raw for nearly all my digital-camera images. But my software collection also includes Canon’s two raw converters, Raw Therapee, and the discontinued Pixmantec RawShooter Essentials (my preferred converter before I got Photoshop). VueScan, my film scanning software, incorporates dcraw code so it can also process camera raw files.

These pictures illustrate how a collection of raw converters can be very useful. They’re a cropped section of a picture of the Carson Mansion, a famous Victorian house in Eureka, California. This image has some kind of unusual configuration of lines that fall slightly askew on the sensor mosaic, and possibly some fine details at the limit of the camera sensor’s resolution. Adobe Camera Raw 4.4.1 rendered it with noticeable red and green striped artifacts on the carved moldings. Although the colored stripes themselves are visible only at pixel-peeping magnification (these sections are 75% of full size), they give edges a strange jagged appearance in normal-sized views or prints. I haven’t noticed this problem on any other raw files, including other pictures of the Carson Mansion.

I ended up using Canon DPP 3.4. Although DPP’s version also shows some very faint striping, it’s noticeable only on very close inspection at full magnification. Without the red and green artifacts, it isn’t a problem. But DPP lacks ACR’s adjustments for highlights, shadows, noise reduction, chromatic aberration, and de-fringing; so I had to spend more time in Photoshop getting the final image. (DPP does have a Curves adjustment, as well as handy presets for color rendering and “picture styles.”) I didn’t see the stripes when I tried the file in RawShooter Essentials and Raw Therapee as a test after I finished working on the picture; so I have to assume that the artifacts are specific to ACR’s demosaicing algorithm. The ACR version looks slightly sharper than DPP, although I disabled sharpening in both raw converters and used the same setting in Focus Magic for “capture sharpening.” (RawShooter Essentials produced the cleanest and sharpest rendering. It resolved the small indentations in the molding around the stained glass window with no sign of stripes or other artifacts. It’s most unfortunate that Adobe “vaporized” this innovative software when they bought out its developer.)

Kuau Sunset is another case where DPP came to the rescue. The Hawaiian sunset scene exceeded the dynamic range of the camera sensor. The clipped highlights at the top of the picture produced ugly artifacts in ACR 3 no matter how I adjusted its processing (the current version 4 might do a better job, but I haven’t tried it). RawShooter Essentials wasn’t much better. But Canon’s DPP rendered the clipped highlights without the artifacts, rescuing this rather nice picture from the recycle bin.

I wouldn’t use DPP regularly, but I’m certainly glad to have it around when ACR doesn’t deliver. Whatever failings camera manufacturers’ software might have, the proprietary knowledge available to its developers may help it handle images with difficult “boundary conditions” better than third-party software, whose developers have to painstakingly reverse-engineer a limited number of raw files. I have had no occasion to use ZoomBrowser EX , but I have no doubt that some day I’ll have a raw file for which its Raw Image Task will produce the best results.

Is A Digital Camera Worth Getting?

The only possible answer to this question is “it depends.” If you shoot a lot of pictures with film, what you save in film and processing could possibly pay for the equipment pretty quickly. If your pictures end up in a digital format— like a Web site, e-mail, or making your own digital prints— a digital camera avoids all the hassles of scanning film. If you travel, a digital camera avoids problems with x-rays at airports (and increasingly at the entrances to public buildings); you also won’t run out of film. And if you like instant gratification, there’s no substitute for seeing your pictures immediately on the camera’s display, or on your television screen.

Conversely, if you don’t shoot many pictures the cost of any sort of digital camera may not be worth it. Film and processing are still be the most cost-effective option, at least for now. If you have a limited budget, a good film SLR (or even a good point-and-shoot) costs a fraction of the price of a digital camera with comparable features. Similarly, if you aren’t handy with computers and software, you won’t be able to do the “post-processing” of digital camera files necessary to get the most out of them. Also, if you can’t stand the certainty of your equipment becoming obsolete in a year or less, a owning a digital camera (or a computer, for that matter) may be unduly stressful.

Since all my photography had been “hybrid digital” since 1999— shooting negatives, scanning them, and doing everything from then on digitally— it was only a matter of waiting until the appropriate digital camera and associated technology became available at a (relatively) attractive price. So the transition, while not entirely smooth, was a relatively straightforward one for me. Someone who has been shooting film only for slides and prints will probably have a more difficult time.