|
When I got my first digital camera (a Nikon point-and-shoot) after many years of shooting film, I was quickly impressed by the resolution, the instant feedback of image review, the gratification of not having to wait for slides to come back from the lab, and the savings in film and processing costs. However, I wasn't too thrilled with the color rendition of the images -- they seemed a little too laid-back and subdued. Actually, this didn't surprise me all that much; from my experience in digital cinematography, I'd come to expect that "digital color" was a little low-key.
Eventually, I learned to start shooting in RAW mode (which is a subject for another article), and, instead of running the RAW through Nikon View's automatic JPEG converter, started processing files myself using Photoshop's Adobe Camera RAW Converter (ACR) . There I found the Saturation control, and found that boosting that value gave me back some of the rich colors I had been missing from film. Generally, an additional 10 percent worked wonders on almost all of my photographs; and 20 made many of them "sing." For particularly drab images, even more saturation could help salvage them. Of course, there would come a point where saturation would cross the line into "too much," rendering images that looked straight out of a '60s-era bad acid trip. Moreover, I encountered the first sign of "photo snobbery" from writers in magazines and on websites, complaining about amateurs always oversaturating and making it seem like any movement away from the "0" point on the saturation control was the sign of a rank beginner, as well as a form of cheating -- using computer artifice to improve mediocre photographs. Was I cheating? Could a photograph only be considered good if not artificially modified? Is there something ultimately dishonest about using the saturation control? But, if so, why did digital photographs often seem, well, bland without using it?
I was stymied for answers until I happened upon Dan Marguilis's outstanding Photoshop LAB Color - The Canyon Conundrum and Other Adventures in the Most Powerful Colorspace. Near the end of chapter 1, Marguilis explains the notion of simultaneous contrast of colors .
According to Marguilis and others, one of the key evolutionary traits humans have developed is the heightened perception of colors when necessary for survival. After all, when humans were living in the wild, they needed to be able to distinguish that tiger creeping up on them through the undergrowth! Although Marguilis doesn't say it himself, it would appear that this heightened perception is something that is triggered by adrenaline, the "fight or flight" hormone that kicks in when one is in a particularly intense situation.
The key point, for photographers, is that the very act of going out to take pictures puts one in such an intense situation. Even if we are stalking a perfect mountain wildflower instead of a bison needed to feed the clan, or if we are "discovering" the Golden Gate Bridge on our summer vacation rather than a mountain pass leading to a new home for one's tribe, our evolutionary heritage ensures that the adrenaline will be flowing -- and that we will be seeing colors as much more vivid than we would at more placid times (such as, say, while looking over a folder of JPEGs on the computer in our library).
Of course, our digital camera runs on batteries, not adrenaline. It doesn't care if a tiger is stalking it! So, when it takes a photo, it records the actual, objective colors that are there, without making them more vivid the way the human mind does. Is that a problem? No...and yes. If your goal is to record the experience you, the photographer, had while viewing that scene, complete with adrenaline-fueled colors, such a photograph is bound to be a disappointment. More to the point, when others view that scene "in the wild," they too will see it through the human experience of simultaneous contrast. What they see will be much closer to what you, rather than your camera, experienced. Therefore, it seems to me quite legitimate to tailor your image to recreate the subjective experience that an average human (like you) would likely have when confronted with the same scene.
You may be thinking that this is all very well and good, but why didn't we experience the same phenomenon back in the days of film photography? There's a simple answer: because films were designed to boost color! Not all of them, certainly, but the emulsions used most frequently for color nature photography rendered colors more saturated than in real life. When I first began in photography, Kodachrome was Kodak's "vivid color" film, while Ektachrome was more laid-back and "natural." Anyone care to guess which one was used primarily by landscape and wildlife photographers? Later, when Fuji introduced the even-more-saturated Velvia, it generated a mass migration of nature photographers more akin to a stampede, while its more-"natural" sibling Provia barely caused a ripple. Whatever purists may say, color photographers have always favored the saturated over the more-neutral. And why shouldn't they? After all, it's what they saw when they were taking the picture!
Now, obviously, there is such a thing as too much saturation in a photograph. Aside from the "day-glo" effect from overdoing color, once you boost saturation in an RGB mode, you eventually start affecting luminance as well, which can quickly degrade an image. That is why, when using Photoshop, I tend to follow Marguilis's lead and convert the image to Lab mode (which decouples color from luminance), then adjust saturation by steepening the curves of the "a" and "b" channels. Going into detail on this technique is beyond the scope of this article; however, an excellent brief tutorial can be found here.
Nowadays, though, I'm not sure you need to process an image in Photoshop in order to successfully tweak saturation. Thanks to Adobe, the 4.x versions of Camera RAW (whose functionality is also included in Lightroom) have a whole series of new color controls. Aside from the possibly-heavy-handed Saturation slider, ACR 4.x comes with a new "Vibrance" control that bases its effect on how saturated the color already is. In other words, subdued colors will be affected the most, while already-bold colors will receive very little tweaking. (Vibrance also excludes the colors that are usually found in fleshtones.) Using Vibrance, you can increase color saturation without fear of pushing bolder colors, or human faces, into the realm of unbelievability. Not enough for you? ACR 4.x also contains individual sliders to let you adjust the hue, saturation, and luminance of each of the primary and secondary colors by itself. The ability to pick one particular color out of an image and tweak it to near-perfection is something you need to experience to fully understand its power. (And, of course, the beauty of using ACR is that the basic image itself isn't being changed, merely the metadata attached to the photo that tells Photoshop how to "develop" that image -- you can change it at any time.) With all these controls, the only reason I would see for needing to go into Photoshop for color work would be if you wanted to mask off a section of the image and adjust the colors for that section alone. (And I understand the upcoming Lightroom 2 will allow at least limited adjustment of selective areas as well.)
Finally, I would note the recent re-emergence of "don't tweak saturation" and even "saturation boosting is cheating!" rhetoric from some photographers (who shall remain nameless) writing for current photo publications. Curiously, these photographers just happen to be the ones whose own images always seemed to me to betray the heaviest hand on the saturation controls. If any of these photographer-writers wish to take issue with what I have written here, I offer them the following deal: If you want to convince me that you can get your results without boosting saturation, let me select a given photograph of yours, and send me a copy of the actual RAW image file for that photo. (To protect your copyright, you only need to send me a cropped portion of your image, not the whole photo, as long as I get to select which portion is used.) Let me see what a "straight" conversion looks like, and whether it bears any resemblance to your published version. If you can really get that result without boosting saturation, I'll offer a full apology and join with you in preaching the "minimal saturation change" gospel. If not, I'll post both the "straight" RAW conversion and your final image side-by-side, and let readers judge for themselves. Deal?