|
If you are using a point'n'shoot camera, deciding on your photo's file format is probably very simple. Your camera will turn out JPEG images. You will probably get a choice of image sizes (how big it will be in pixels), and "quality" (how much compression will remove of the image data that, theoretically, you'd never notice in the first place), but that's about it. If you have a slightly-more-expensive camera, you may have the option to save in "TIFF" format instead. This, unlike JPEG, doesn't compress the image, so you get full resolution, at the cost of a file that is probably three to five times larger.
But, in either instance, your camera is processing the image in the same way -- it's taking the reading from the red, green, and blue sensors on your imaging chip, using "demosaicing" algorithms to turn those red, green, and blue readouts into full-color pixel values, applying whatever image choices (contrast, saturation, etc.) you've set in user preferences, and writing the image out to the memory card as a JPEG or TIFF file.
If you have a D-SLR, or one of the top-of-the-line point'n'shoots, you've got another file option: something called "RAW." What on earth is RAW? Well, to give the simplest possible answer, it's a way of storing an image that is unlike any of the others mentioned above, and truer to your original capture.
Remember all the in-camera image processing steps I mentioned above? Well, if you're shooting in RAW, your camera doesn't do any of them! Instead, it simply writes out the readings from those red, green, and blue sensors, either using a proprietary format or, in some cases, Adobe's universal "Digital Negative" (or DNG) format. When you download the images to your computer, you open them in a paint program such as Photoshop or a stand-alone RAW converter such as Bibble Pro, and process the image to your own liking instead of what your camera assumes you would want. You can set the white and black "endpoints" of your image to get greater exposure latitude, set the brightness and contrast the way you want it, dial up (or down) the color saturation, even adjust the white balance. When you're done, the converter will either open it up in the paint program (for saving in whatever format you want) or save it as a TIFF or JPEG file.
Why would you want to go through all the trouble? Well, as they say, a picture is worth a thousand words but, before going to the pictures, let me throw in a few dozen words of my own. When I started out, I had a Nikon 5700 that was one of the few point'n'shoots that offered RAW. Since I had heard it was the best choice for flexibility, I shot in RAW, but simply downloaded the files to my computer and had Nikon's in-the-box-with-the-camera Nikon View software automatically turn the RAW files into JPEGs (fortunately, as it will turn out later, I saved both the RAWs and the JPEGs). In other words, I was letting Nikon View do exactly what would be done in-camera otherwise -- take the sensor data and automatically generate a JPEG file without user input.
For some time (well, actually, a couple of years), I was quite content with that arrangement. Then, one June day in 2005, I went with a friend on a shoot to Deception Falls State Park. I took a number of photos of Deception Falls under just about the worst lighting conditions possible -- bright, glaring sunlight. I then automatically turned the files into JPEGs with Nikon View.
The results, as you can see, were not all that good. Because of the brightness of the highlights, the shadow areas of the rocks were pushed way down into almost pure black. But even worse was the color, or near lack thereof. Deception Falls is a very colorful place, with moss-covered rocks and fir trees lining the sides. But the image I'd taken could just as well have been in black and white. Even more embarrassing, my fellow photographer had come up with images that, while still suffering from the extreme contrast, at least showed a healthy amount of the color that was so totally lacking from my own images. (And, while I must say I like a good black-and-white image as much as the next photographer, it helps if the image really was intended that way from the start!)
Well, after wondering about the situation, for some time, I began to think about the fact that my images were in RAW, and thus theoretically should have had more information than my friend's JPEG-from-the-start images. So, I finally got the impetus to open that image in Photoshop's RAW Converter, and try adjusting the image settings before exporting it to JPEG. Since I had little experience in RAW conversion, I merely experimented with the most basic functions such as black and white point setting, brightness and contrast, and color saturation.
Need I say that the difference was not subtle? This time, Deception Falls appeared to be in color! Not just the tree branches and moss on the rocks, but even the water carried a touch of cool blue in the shadow areas. And there was a lot more detail in those shadows, as well as in the mid-tones. Processing the RAW myself turned a truly bad image into a rather evocative one. And that, of course, only raised the issue of "how much could I gain from going back and reprocessing all of my images of the past two years?" As it turned out (a few weeks of Photoshop work later) the answer was "a lot." Not to bring religion into it, but it was as if my old photographs had been "born again."
Unfortunately, I don't have any of the old, auto-processed JPEGs around except the one above. But have a look at the two of them again, this time at a much larger size:
Since then, needless to say, I've made a point of hand-converting every photograph I take (including all the work you see on this site) from RAW. One of the greatest benefits (which isn't seen in the comparison above) is that shooting in RAW and processing it later often allows you to "recover" highlights, as most RAW converter white point setting controls will restore what initially seem to be blown-out whites and actually find hidden detail and texture in them. (The reason you don't see it in the photo above is that Nikon View seems to have concentrated on avoiding blown-out whites as much as possible, at the expense of forcing shadow areas to black. For example, look at the pure black of the large rock at the left in the auto-converted image, then compare it to the shades of dark texture in the manually-converted image.) While it doesn't replace attention to getting the exposure right in the first place, RAW conversion has the ability to rescue an image taken in a particularly difficult, contrasty environment.
And there's even more reason to shoot in RAW now, because of Adobe's fantastic Lightroom software. If you shoot in RAW, you can have Adobe import it (and, better yet, convert it to their DNG format, which preserves all the proprietary RAW data) and catalog it that way. Then, when you process the image, you aren't changing it at all, merely having Lightroom's own catalog record a history of details on how to convert the RAW or DNG file when printed or exported. Lightroom's conversion controls (the same as those found in the new CS3 RAW converter) go far beyond those I used to convert the image above -- in addition to the simple brightness, contrast, saturation, etc. controls, there now are adjustments to "vibrance" (boost only unsaturated colors), "fill light" (boost the darker parts of the image), and even a way to set brightness curves and specific colors by clicking on the appropriate point of an image and scrolling your mouse up or down. The number of image processing tricks you can perform on RAW/DNG files in Lightroom is almost unbelievable...and, since you're only recording changes to the catalog, your stored image itself is never changed at all, meaning that you can go back and re-process it at any time, to your heart's content.
I know there are those, including some well-known photographers, who consider RAW too much of a "time-waster," and prefer to just have their camera write out TIFFs and JPEGs. And I suppose, if they know their preferences well-enough and are able to set their cameras' presets clearly-enough, they might be able to get equivalent results...most of the time. But, before you decide to emulate them, I recommend that you try shooting in RAW for awhile and practice converting your own images. Or, better yet, since most DSLRs have a setting where the camera saves the image as both a RAW and a JPEG, try shooting that way for a time, then convert your RAW images and compare them to the camera-generated JPEGs of the same images. I suspect that, if you really value the quality of your images, you will likely be shooting RAW from then on.