The RAW Format

The Difference Between JPEG and RAW Photography

Shooting RAW capture versus shooting JPEG is to a digital photographer what making a cake from scratch versus making one from a mix is to a baker. The mix cake tastes good, looks good, saves a lot of work for the baker, and meets most needs. Starting from scratch, though, allows for the baker’s own special touches and as much creativity as he or she wants to exercise. A cake mix is good for kitchen rookies, because it doesn’t require as much decision-making as the customized cake. Those with more confidence, who understand the difference between baking soda and baking powder, may want to make their personal time with the mixer a one-of-a-kind culinary adventure.

Shooting JPEG is what new, recreational, and sometimes professional digital photographers do. Digital cameras come with built-in converters that take the data received from a photo shot and manipulate, then compress, it into a JPEG file, making the production of quality photos as easy as framing the shot and pushing the button. Camera makers know that the only way to bring photography to the teeming masses is to make photo production easy. And, indeed, most of us are satisfied with the way our digital photos look as JPEG files.

Shooting RAW gives the photog exactly that: unmanipulated raw data that lets the shooter bypass the automatic data conversion process. Partnered with a RAW-ready digital camera, RAW data converter software allows shooters to have more control over the final result by allowing for any number of personal decisions about how to achieve it.

How Digital Photography Works

To appreciate the differences between RAW and JPEG, it helps to know how a picture is made. Let’s say that Skip and Chip are standing side-by-side, identical cameras in hand, looking out over a field of snow-covered corn stubble currently occupied by a flock of wild turkeys. Each of them raises his camera, frames the shot of said turkeys, and snaps.

Everything that Skip and Chip do looks the same so far, right? It is.

But Skip, being older and more worldly, uses a RAW-capable camera, while Chip’s camera shot in the popular JPEG image file format. Each camera is storing the same two kinds of information: image pixels and image metadata, which is simply data about the data, including information about the model of camera used, shutter speed, aperture, the date the photo was taken, and other camera settings. Chip’s camera automatically converts the files according to settings determined by the camera maker, tossing a great deal of the captured information to the curb, and compressing what’s left into the JPEG image file format. So his work is done, except for some minor editing if he wants it and downloading the photo onto his computer or phone or other 21st century techno-gadget.

Skip wants his turkeys to have deep color tones and smooth edges. His work has only just begun.

Developing Skip’s RAW Turkeys

The only settings that are preprogrammed on Skip’s camera are ISO, shutter speed, and aperture. While his camera has captured the same image pixels and image metadata as Chip’s, it also provides additional metadata necessary for him to successfully convert the RAW information to a color image. Elements of conversion require decisions about the following:

• Color, or colorimetric interpretation: Digital cameras capable of performing RAW capture commonly use what is known as “mosaic sensor” or “color filter array” — CFA — technology. Arranged in a grid, photo sensors provide pixel data that is converted to an RGB — red, blue, and green — image through a process similar to “developing” film, called “demosaicing” (a strange-looking word that defies the laws of phonics — it looks like “demo-sasing — but includes the root word “mosaic” if anyone needs help with its pronunciation). Each sensor produces one pixel and is almost always overlaid with a “Bayer” filter, making it only red, blue, or green, each of which converts the RAW grayscale image into a specific shade of the color, depending on the light level. There are also filters based on the CMY scheme — cyan, magenta, and yellow — and some that allow a fourth color, but RGB is the most common.
   
  RAW pixels have 12 – 16 bits of color information, allowing more subtle color variations. JPEG pixels have eight bits, which is about all the human eye can discern.

• White balance: The photographer can adjust the red, green, and blue shades to ensure the grays and whites are pure and not tinted by any of the colors (called “color cast”).
• Color saturation: The dominance of the color can be maximized; the more saturated it is, the less gray it has.
• Contrast and sharpness: Contrast, the difference between the lightest and darkest tones of a photo; and sharpness, the opposite of blurriness, can be modified to the photog’s liking.
• Noise reduction and antialiasing: There are two kinds of noise, luminance and chroma. Luminance refers to graininess, and chroma refers to mischievous spots of red and blue. Reducing noise can also decrease sharpness and damage color, so it requires a delicate balance. Antialiasing refers to adjustments made to the edges of colors in a photo so that they don’t appear jagged and ragged close up.
• Exposure correction: A common problem photographers encounter is photos that are over-exposed (too light) or underexposed (too dark).

After the adjustments and enhancements are complete, the RAW image can be saved as either a JPEG or a TIFF. The JPEG format requires aggressive compression to an 8-bit image which may compromise the quality of the photo. A TIFF may or may not require compression (to between an 8-bit and a 16-bit image) but its compression is lossless, which means that in the process of becoming smaller, no information is lost and the picture’s quality remains intact.

Why Some Like It RAW

Besides being the obvious choice of control freaks, the fact that RAW development allows more room for corrections and adjustments is one of the reasons its fans like it for some of their work (it would be extremely time-consuming for the non-professional to use RAW capture for every photo). It allows for storage of an original unmanipulated photo, like a negative; and, again, for corrections to under- and overexposures, and its high-bit depth ensures smooth color gradations and allows for more extensive adjustments described earlier.

Why RAW Photographers are Frustrated

RAW capture doesn’t follow a universal standard. It is camera-specific, a subjective decision made by the different camera-makers that determines the parameters for its cameras’ RAW capabilities. They even use different extensions: For example, Canon’s RAW file extension is .crw, Nikon’s is .nef, and Minolta’s is .mrw. There is concern among photographers about the lack of standardization and, as a result, a growing push for it. Adobe’s .dng is the closest there is to a standard, but making it universal seems to be stalled due to disagreement among camera makers. Photographers are justified in their concern about a universal standard, because different RAW formats can come and go from the marketplace, leaving them with obsolete and unusable files.

Most digital SLR and high-end cameras are RAW-ready, and RAW software, which can be downloaded, is a necessity. Do-it-yourself RAW file conversion is definitely not for beginners, but for experienced photographers like Skip who are bored with JPEGs and ready to tackle their own photo manipulation. In the end, Skip has been able to capture the subtleties and shadings of his turkeys — the vivid red wattles, the bronze iridescence, the striking black and white wing bars — much like a portrait photographer. On the other hand, Chip knows his turkey photos won’t be enlarged so they can be ogled in a coffee table book, so he, too, is happy with his JPEG downloads, which are certainly suitable for framing.