Most modern digital SLRs can shoot in multiple file formats. The ubiquitous JPEG (an open-source standard from the Joint Photographic Experts Group) format seems to be the defacto standard for working with digital images (in fact, almost every image file online is a JPEG-format image), and pretty much every digital camera out there captures in JPEG. JPEGs store images based on the RGB values of each pixel, and use specialized compression algorithms to lower the file size—at the expense of quality.

There is also the TIFF format (it stands for Tagged Image File Format). TIFFs are somewhat of a standard for printing—they support CMYK color modes, layers, transparency and lossless compression (meaning that unlike JPEG compression, no data is lost. TIFFs can be very large files and outside of the graphic design industry, are not easily shared or used.

Digital SLRs have a third option: Camera RAW. A RAW file is proprietary to the camera manufacturer and uses a custom, secret format to store all of the information captured by the camera’s sensor. This is the highest quality file format, but it can be a bit confusing—terms like “CR2″, “NEF”, “12- or 14-bit RGB” are often poorly understood, and the hassle of sharing or viewing the wierd files that come out of the camera can be frustrating. In this article, I hope to clear away some of the mysteries surrounding Camera RAW and convince you to switch to RAW as your primary shooting setting. Click “More…” to keep reading!

What happens inside your camera, stays inside your camera…or not

When you take a picture, a lot happens inside your camera. Light comes in through the open shutter, and bombards the sensor. Millions of tiny, color-coded sensors measure the cumulative amount of “photons” (light particles) that hit it. The shutter closes, and the data-value of each pixel is rapidly drained through a series of “pipes” that lead into the chip circuitry. A specialized chip, known as an “Analog to Digital Converter” assigns a numeric value to each pixel’s data measurement. At this point, a set of expensive image processing chips take over and turn this raw database of pixel values into a finished JPEG image. Factors like white balance, exposure compensation, quality setting, sharpening, lens correction, color mode, and custom processing settings are all applied to the data, and the camera then written to the camera’s storage card. This whole process is completed in a matter of milliseconds, and as much as 98% of the data generated by the sensor will be thrown away by the time the final image hits the card. This JPEG file is usable, share-able, and finished. And once it is taken, there is very little you can do to fix it in post-processing—adjust any of the parameters outside of a very narrow safe zone, and you will experience unacceptable quality loss.

For consumer digital cameras, this process is accepted and unchanged. For digital SLR users, however, there are more options. A Camera RAW file takes all of the data from the Analog to Digital converter and streams it, completely raw, unedited onto the camera card. It then adds data about white balance, and lens corrections. It even generates a “preview” image that looks like all of the camera processing settings. But the original data is all there in case you decide to change something later on. As a result, the file is a lot bigger, and it is unfinished—in order to share it with anybody, you need to process it into a JPEG later on.

suddenly color and light are a lot more complicated.

JPEGs support an 8-bit RGB color space. Let me explain that: If you were to picture all of the colors in the world, laid out on a graph, a color space is the section of those colors that can be displayed by a given color mode. It is also called the “gamut” which refers to the range of colors that can be output in a given mode, or on a given device. RGB stands for Red Green Blue and means that for each pixel, a value is assigned to the red component, a value assigned to the green component, and a value assigned to the blue component. Remember how you were taught the primary colors in elementary school? You were taught that they were Red, Yellow and Blue. I remember mixing paints in those colors—by mixing Red and Blue, you got purple. Yellow and Blue made green, and so forth. Those are the primary “reflective” colors. Light, however, behaves differently. Because it is transmitting light, the primary colors of light are actually red, green and blue. And so the JPEG image file shows each color as a combination of three numbers. Each number takes up 8 bits of space. A bit is the basic unit of information. One bit is equal to a one-digit binary number and can have only two value—on or off, 1 or 0. Two bits can have four value: 0, 1, 10, or 11 (equivalent to 0, 1, 2, and 3 in decimal numbers). An 8-bit number can represent 256 values. Thus, an 8-bit RGB color space represents each pixel of light and color as a combination of values from 0 to 255 for each color.

Most camera sensors can capture significantly more light data at each pixel site. Thus, they have higher “bit depths” (put another way, they can be more precise in storing light data). A 10-bit sensor can describe each color value as a number from 0 to 1024. A 12-bit sensor describes it as one of 4,096 values. A 14-bit sensor actually can generate 16,384 values! That is 6400% more precise than an 8-bit image.

Camera RAW files will store the full bit-depth of an image, so the quality of information is much higher. Of course, the files are a lot bigger too.

CR2, NEF…what’s with the funky files?

Each camera’s internal computer chips process the data produced by the sensor differently, and so each camera manufacturer has come up with their own custom file formats to store it. Canon cameras create RAW files marked “.cr2″, and Nikon RAW files are labeled “.nef.” This can be frustrating, because these files are not easily opened in every software program. The future is even bleaker—because they are “closed” file formats (meaning only the manufacturer knows how they work) these files could be completely unreadable if the camera company goes out of business, or changes formats, or stops supporting the format.

Fortunately, the Adobe software company has come to our rescue. Many years ago, Adobe created the “PDF” (Portable Document Format) as a an open-standard file format to share documents. Now, it pretty much a global standard for sharing documents. Because it was created as an “open” format, any one can build software to create or to read PDF documents.

Adobe has recently created the “DNG” standard in the same way. An acronym for “Digital Negative”, this format is basically an open RAW format. It is designed to hold all of the RAW camera data, but can also store processing instructions and metadata (such as keywords and camera information) in the open-standard XMP metadata format right within the file. Fifty years from now, if Adobe were to go out of business, people could still write software to open DNG files and use them. It is designed to be the ideal way to store photographs long term at high quality.

A few cameras actually will shoot in the DNG format, but for everyone else, you must convert them first. Fortunately, Adobe has a free DNG converter available from their website. This will easily take your camera’s RAW files and turn them into DNGs. For those with Adobe Photoshop, the included Adobe Bridge CS3 can also convert to DNG.

Personally, I use Adobe Photoshop Lightroom to manage my photos, and it can actually convert my RAW files into DNGs when I import them from the camera. For me, this is the best option of all. It does slow down the import some, but I know my images will be secure long-term. I can also convert my RAW files into DNG within Lightroom.

For more information on why DNG is such a great choice, I highly recommend this video by Terry White from the Adobe Creative Suite Video Podcast:

post-processing raw files

Whether you are using your camera’s RAW format or Adobe’s DNG format, post-processing is the same. If you are using the Adobe software, you can use Adobe Camera RAW (a software program) or Adobe Photoshop Lightroom to make the desired adjustments. You can often make the same adjustments in the software that came with your camera.

In a film-based photographic workflow, significant time is spent in the darkroom developing the film and making the prints. Serious digital photographers should place the same importance on post-processing their images, and should make the “development” of their RAW files an integral part of their workflow.

I am going to demonstrate my processing workflow in Adobe Camera Raw 4.3.1; the same adjustments are available in Adobe Photoshop Lightroom.

When I imported this batch of files off of my camera card, I had it automatically convert the NEF RAW files into DNGs. I then opened this one using Adobe Bridge into Adobe Camera Raw.

Several things to note in this initial view: 1) this toolbar contains the key tools you will be using (much like in Photoshop). 2) The histogram is a key tool for understanding exposure. In both Lightroom and in Photoshop, this also shows your c

amera exposure settings. 3) These are the different “tabs” of sliders. In Lightroom, each of these tabs is instead shown as a module. 4) This section contains the sliders we are concerned about. 5) A little bit of metadata (literally, “data about the data”): The color mode (Adobe RGB(1998); unless you have a color-managed workflow, you should be shooting sRGB not Adobe), the bit depth (8-bit), the image size and megapixels and the printing resolution (which you can ignore).

In post-processing my images, I generally try to follow the order of the tools in Adobe Camera RAW or in Lightroom: Left to Right and Top to Bottom.

The first thing I will usually do is cropping. I really hate to crop, because I want to get it right in camera, but I personally feel that the post-processing is as much a part of the photography process as the actual shooting. That is how I justify my computer compensation for not getting it right in the camera. There are other perspectives on this, and if you want to explore the ethics surrounding this, Scott Kelby’s article “My Photo Editing Code of Ethics.”

When cropping, remember the rule of thirds and think carefully about composition. You can also crop to a specific aspect ratio; often when taking the photo I’ll visualize it as a 5×7 or 8×10 frame, not the default 4×6. You can select a custom aspect ratio to use for cropping by pressing and holding on the Crop tool icon, or in the Aspect ratio option box in Lightroom.

At this time, I will also straighten the image, remove spots and fix red eye (that is the order of the tools, from left-to-right).

The next step for me is adjusting the White Balance. For set-up, studio shots, I’ll do this in-camera using a white card, and I usually don’t have to adjust it later. Most of my photos are done on the fly using the equipment on me or in my bag, under time constraints and so I have to finesse the white balance in post. This is easily done using the presets (use them just like in your camera). Changing the preset here to “Daylight” gave me pretty pleasing results. For further tweaking, you can adjust the sliders for temperature and tint. If you have a neutral colored white or gray object in the frame, you can also use the “White Balance” tool (the eyedropper at top) to quickly and easily nail the white balance.

Once the white balance is correct, I’ll adjust the exposure settings. Some of this is subjective, but for much of it, I use the histogram at top as a guide. The histogram graphs the TOTAL amount of pixels in the image at each light value, with the darkest value at the left and the lightest value at the right. This gives a really good indication of the distribution of light values in the image. In this image, much of the content is in the shadows. This is the raised part of the histogram on the left. The leaves and flower are middle-toned and are represented by the little “mound” towards the middle of the histogram. With digital, you “expose to the right”, meaning that you want the histogram to be pushed as far towards the right as possible without “blowing out” the highlights. Since there is a limited range of light values that can be shown, anything brighter than the brightest light value is automatically cut off. A peak pushing off the right indicates a blown-out highlight. These areas of the image lose all detail and become these ugly white blotches. Hardly desirable.

I start with the exposure slider, moving it until the histogram roughly resembles what I want it to. This can be somewhat subjective; I use a variation of Ansel Adam’s Zone system that is quite complicated; a little too much so for this article.

At this point, some of the highlights are blown out, so I use the Recovery slider to “recover” those highlights—basically the recovery slider acts on the highlights only and lets me pull those back down. I used the Fill Light slider to bring up the shadows (it is basically the reverse of the Recovery slider), and then used the Blacks slider to add a little pop to the image. The Blacks slider pulls down shadows and clips them, which results in richer blacks (although you lose shadow detail in the process).

I typically leave the Brightness/Contrast sliders pretty much alone. The Clarity slider is important. It acts on the edges in an image to emphasize the textures, patterns and shapes. I love it. The Vibrance slider is another key. I like rich, vibrant images and the Vibrance slider does a good job of boosting the colors without oversaturating them. The Saturation slider is rarely used unless I really want to boost the saturation of colors—or lower them for creative effect.

The next step for me is the Tone Curve. The Tone Curve is a pretty sweet invention that allows you to define a graph to process the image through. Basically, the horizontal axis is the input access. Like the histogram shown in gray, light values on the left are the shadows and on the right are highlights. The vertical axis is the output values (with light at the top and dark at the bottom). A straight line represents a strictly linear response; at any point along the curve, the x value (input) is equal to the y value (output). I like to use the curve add gentle contrast. In this case, the point at the top means that any pixel with a light value of 173 will be boosted to 194. Likewise, the bottom point shows that any pixel with a light value of 37 will be dropped to 25. The curve graphs this response out smoothly across all of the light values.

The final thing I do on most images is sharpening. For this, you’ll want to zoom in all the way to 100%. The sharpening sliders are unique in that you can hold down the ALT (Mac:OPT) key while adjusting the value to get a real time preview of the effect. The Amount slider is pretty self-explanatory. The Radius slider basically controls how sharp the edge ends up being by controlling how wide of a radius is sharpened around each edge. The Detail slider adjusts how sharp the non-edge areas are, and the Masking slider limits the sharpening effect to edges only.

In a really noisy, high ISO image, I’ll also use the Noise Reduction sliders. Having been shot at ISO 200, they really don’t help this image any.

At this point I have a fairly finished color image. I can save it and it saves all of my adjustments as “instructions” inside the DNG file. At any time, I can reopen the image and adjust it further or reset all of the adjustments.

Most of the time, this is as far as I go with an image. I may open it up in Ph

otoshop and tweak it a little bit, but typically I don’t have time to go that far.

If I were to make this image black and white, I would also be using the HSL/Color/Grayscale panel and the Split Toning panel. (Which I’ll cover in another article).

This process may seem long and involved, but remember, if you are using Lightroom, you can easily syncronise your settings across an entire batch of photos.

In Lightroom, these adjustments are available and mostly the same for your JPEG images. The amount of available source data is much less though. In order to get the highest quality output, you need the highest quality input—Camera RAW.

Downsides to shooting in raw

I hope that you will start shooting in RAW more often now that you’ve read this article and seen all the advantages. But, RAW is not a perfect solution. For starters, RAW files take up a LOT of space. On my 2GB CF card in my Nikon D200, I can fit 658 full-size, full-quality JPEG images. I can only fit 119 RAW images. This means that I can get less on a card, and that downloading that card onto my computer takes a lot longer as well. Additionally, there is a small impact on performance when shooting, since the files take longer to write to the card in camera. Shooting in RAW also means you can’t plug your camera into someone’s computer and share your photos with them. Many cameras have a RAW+JPEG option, which gets around this by storing two copies of the image on your card. You can easily share the JPEG file with someone, and the full-size RAW is available for later use. This is useful in other settings as well—for instance, if you have a tight deadline to get photos someplace, you can import and share the JPEGs; but you’ll still have the RAW files if you need them. The downside to RAW+JPEG is that it takes up even more space on the card.

I shoot RAW for almost everything. I’ll shoot JPEG for events, where I may end up taking several hundred images and I don’t need to do a lot of post-processing high quality work on them.

Let me know in the comments what modes you shoot in and why!