A Tale of Two Cameras

This is a set-up. I’m going to tell you a short story and raise some questions; then I’d like to know what you think. It won’t be as easy as it might seem but it should be fun, and maybe we’ll both learn something useful.

I’m currently using two cameras on a regular basis, and one takes prettier pictures than the other. This might not be a surprise, as one is a cheapie, bottom of the line backup that I keep with me to handle the minority of what I do and to be there if something happens to #1; it was chosen because it was inexpensive, light weight, and handles all the lenses the other camera uses. It also takes quite credible pictures, but is programmed to wear out and depart long before #1 begins to breathe hard. 

The main camera, #1, cost an arm and a leg, features a few K more megapixels, is weatherproofed, has a shutter that will exceed 10,000 gerbil lifetimes, and is big enough for a family of those little rodents to call home. It is also heavy and harder to live with, doesn’t have the frequently useful (but lacking in “pro image”) built in flash of its backup buddy, and can cost more to repair than the replacement cost of #2. 

So why is it that it is the cheaper camera that takes the prettier pictures? Simple. The big guy has been told that it is not to enhance contrast, not to augment color, and under no circumstances to try to sharpen anything. It is to work in a color space called Adobe RGB rather than the more common and vivid sRGB. Bottom line: a digital file from camera lite could go directly to the friendly photo printer at the drugstore and make great prints for the family or the internet; the heavyweight could embarrass me if I tried to do that, because its files must be interpreted by a human person (me) before they are fit for general consumption. 

Which brings us around to my point. Modern digital cameras don’t take pictures (or “capture images” to use the fashionable phrase of the moment). What they capture is electronic signals that are then processed, both onboard and after the fact (read Photoshop or some other imaging program). What I want to see is a “raw” file with all my options still open; I’d rather do the final interpretation than have an on-board algorithm do it and then toss out those considerable bits of info it deems unnecessary. End of the day, everything is interpreted one way or another; there is no absolute gold standard of this-is-the-way-that-it-was when the capture was made.

Here’s the question: how much interpretation is fair? This covers a lot of ground, from adjusting color temperature, to contrast, to lightness, to actual physical distortion of the image and removal or addition of some features, even into the land of the 2-headed Martian babies beloved by the tabloids. How do you feel about this? I’d like to know, and if you’d like I’ll share my personal standards of practice with you at a later date. Meantime, drop me a note: leonardt@pca.org.