Sunday, July 12, 2009


As happens every so often, I was taken to task for presumably denigrating an individual’s photo activities by my use of the term “snapshot’ in reference to the on-line services that provide inexpensive printing from JPEG files. Although there can be considerable crossover between snapshots and photographs, the pictures serious photo enthusiasts refer to, in my perspective of things, is not a value or status distinction.

Some years ago I believe it was Sony that had a billboard advertising a new video cam in which video was likened to “moving snapshots”. I thought that astute and apt, because most people who make personal home video do so for a social purposes, and I believe that is the purpose distinction that differentiates snapshots from the photographs photo enthusiast make. A photographer’s purpose in making images is to create photographs for their own sake, or as part of an interest in taking pictures of animals, flowers, landscapes or auto racing. In other words, what distinguishes snapshots and photographs is NOT that one is better than the other, or the person behind the camera is superior or inferior, but the purpose in using a camera in one case is social, about people in the picture taker’s life, or events like family trips or an outing at the park, a day at Disneyland; while the purpose of making a “photograph” is the image, the picture that is created and sometimes what is involved in the process of image creation as well.

In the 30 some years I have been in the photo magazine trade, I have looked at an awful lot of pictures made with a camera, both “photographs” and snapshots. Among them, in both categories, snapshots and photographs, I have seen examples of superb technical craft and skill as well as artful composition and sensitivity to the subject, as well as a multitude of mediocre images in both snapshots and photographs, to really dismal failures as pictures of any kind or description. But there is one distinction the makers of snapshots do have and that is they outnumber serious enthusiast photographers by many times. So of course business tends to favor large numbers when it comes to customers, and most of the print services cater to the snapshooter, while those that cater to photographers are fewer and usually because volume is less charge higher prices.

The same kind of numbers dominance also affects technology. The default settings for most digital cameras, even dSLR’s is to save files in JPEG and the often associated sRGB colorspace (and for that matter computers and image editing software). So likewise the bulk of print services expect or demand image files from customers conform to the JPEG format, and usually their printer driver computer use the sRGB colorspace to print from. There is nothing wrong with that if the customer is pleased and satisfied with the print results, the customer is always right.

But for the fewer, the serious digital photography enthusiasts, one has to go beyond the default setting, and use non-lossy file formats like TIFF and a working space profile like Adobe RGB (1998). These differences cause confusion for novices, and newcomers from film to digital, as well become more bones of contention in themselves and inadvertent complications between the snapshot and photograph crowds. It generates almost as much invective and vitriol as the PC Window versus Apple Mac user debates.

So what is the reality. As I said the customer is always right, if a camera user is happy with JPEG/sRGB results who is to argue. These happy campers, whether they are aware or not accept the price JPEG/sRGB imposes in lost image content data for the conveniences of being a part of an established crowd. But for the record let’s take another look at the reality (and history) of both JPEG and sRGB to understand why they are established and play the role that they do.

First of all, as the web became a significant part of the internet, and photos became content included in web pages, the file size of those images was an issue, which if too big slowed the internet to a crawl. And in those days not that many years ago hard drive and mobile storage space for files were small in capacity and expensive per megabyte, making both JPEG and sRGB because they both reduce file size a double advantage. But some of us (serious photography enthusiasts) argued realistically that the cost in terms of lost image data content and quality was too high a price.

As an example let’s look at each of these two different strategies, JPEG and sRGB individually. The sRGB colorspace was promoted by some like Microsoft, Intel and Hewlett-Packard, not just as a solution to the limited bandwidth of internet transmission, but at one time as a cheaper, simpler, and of course more profitable way to dispense with the need for color management, to get a printer to make color prints that match the color on screen. Those companies did not succeed in getting sRGB to be the international standard for all hardware and software that reproduced color because many powerful factions like printing and publishing, the video and motion picture industries and all the professionals that supported that level of work were vehemently opposed to being put in a color straight jacket. The sRGB colorspace profile is a simplistic solution to achieve color matching, by just reducing the size of the color gamut to the lowest common denominator of what the least adequate color monitor could reproduce, then all color images look the same on every monitor. But the price of that approach is if you convert a dSLR Raw image to sRGB you discard, throw out, a third of the color the image sensor recorded!!!!!

The purpose of the JPEG compressed file format is different and simpler, to reduce file size, but the effect on a photographic image is similar to sRGB. Image information is thrown out, lost forever, and color is simplified. The reason is the strategy that allows reducing the file size is, in oversimplified fashion, to re-write the file information not as individual pixels, but a grid of blocks of pixels, lets say 4x4 pixels or 16 pixels in a block. Then if a block of pixels is part of the sky in a photo, the JPEG processing recognizes all of the pixels in the block are similar, so it says the differences are actually redundant and not needed, so let’s instead of writing 16 lines of code, one for each pixel, write just two lines to say all the pixels in the block of sixteen are the same RGB color value. Depending on whether High, Medium, or Low compression quality is chosen more or less color differences are made to be the same, reducing the variation in subtle color differences in the image.

Camera users seldom complain about this loss because the visual effect is that as color variation is reduced, color looks more vivid and contrast more bold. I call this effect “cartoon color”, because cartoon movies to make each frame in a film consistent in color throughout the movie used a very limited palette of colors to create each hand painted cell that is copied on film to make films like Disney’s now famous animated movies. And like a cartoon color image sRGB/JPEG images with their simpler and smaller color palette are also a step away from reality, like the reproduction of a subject a dSLR camera captures when set to record in Raw format.

Of course the more realistic advantage of greater color fidelity to a subject Adobe RGB (1998) from a Raw dSLR camera file provides has the down side of requiring the individual photographer to color correct, adjust and process the image using an application like Photoshop manually instead of having the camera’s internal auto processing do it for them when the JPEG file save option is chosen.

If you have a comment, they are welcome, so please post it. If you have a question you want me to answer please address an e-mail to David B. Brooks at:

And visit my web site at:

No comments:

Post a Comment

You can also ask questions I will answer privately by sending me an e-mail at: