Tips & Techniques

RAW! why not cook our files?

CS2 RAW Conversion

The basic answer is all files are captured as RAW - but when set to jpeg, the camera does the RAW conversion and etches your parameter settings, white balance, and exposure in stone - then compresses the file into a jpeg, thus partially developing your 'film'. You can then take that jpeg and complete the developing in a program such as photoshop, however you will be chiseling away at that stone, and if you end up fighting what the camera has already done, you will see the artifacts of that battle. Image quality can and most likely will suffer the consequences.

A RAW file comes into your computer with the camera as-shot settings and parameters noted - but not set. You then have the option of changing your mind almost as if you were on the scene again - possibly even better. These changes, within reason, will not leave battle scars on your image such as editing a jpeg can. You are working with clay rather than stone.

Now, why can RAW conversion be even better than the ability to re-shoot your image? Some changes can only be done in-camera, such as focus, and there is no way to correct a mistake made in the field. But for other parameters RAW conversion can be better because there is actually more flexibility and control in the conversion than in the camera.

Take "contrast" as an example. In most cameras, you can set the contrast in something like -2, -1, 0, +1, or +2 increments. The plus = more contrast and the the minus = less contrast. You don't really know exactly what you are asking your camera to do other than increase or decrease contrast. In RAW conversion just play with the contrast slider until you get it just right, or even better, adjust only the shadows or the brightness, leaving the other parameter alone - something that can't be done in-camera without affecting the opposite parameter. If you want to deepen the shadows in-camera you can decrease the light through an exposure adjustment - thus darkening all tones and losing valuable information [that is a topic for a future discussion!], or you can increase your contrast setting, thus potentially blowing your highlights, or losing the shadows completely, or both. In RAW conversion, deepen the shadows without changing how the remaining tones are rendered! Now, can that be done with a jpeg? Yes, to some degree . . . through something like a levels adjustment in PS. Have you ever done a levels adjustment then looked at the histogram in the levels dialogue box? [If unfamiliar with the term histogram it is simply a map of the image tones - from black on the far left to white on the far right] It will go from something like this:

to something like this after a relatively small adjustment:

What are all those lines and spikes in the new histogram? The histogram simply maps the tones in the image - the gaps are tones removed from the original to make the adjustment - scars chiseled into the stone.

Contrast that "after" histogram with this "after" histogram from a similar adjustment made during the RAW conversion:

The histogram, or tonality map, is different than the original as-shot - but largely intact without the gaps or "scars" - the image was molded rather than chiseled.

White balance in RAW conversion is FANTASTIC - not only can you change what the camera recorded for white balance, but you can adjust it to suit your artistic vision.

RAW is not for all shooting in all situations - it fills your buffer faster and uses more memory per image. A sporting event in good light is typically a case for jpeg. A wedding in low, mixed light? Well,if you don't shoot RAW you WILL feel the pain, AND end up spending more time editing with less satisfying results. Trust me - I've been there!

A short primer "The RAW Truth"(c) by Michael Reichmann, for those interested in learning more about RAW file format vs. JPEG/Tif, can be found here: (http://www.luminous-landscape.com/essays/rawtruth1.shtml)
The article links to a more technical paper by Michael for the tech inclined ;). But these tables from Michael's paper sum up the impressive difference in quality opportunity between a RAW and JPEG. The levels referred to here are tones.

CS2 currently offers 16 bit RAW conversion, but it is my understanding it is simply taking the 12 bit data from the camera and spreading it over a 16 bit depth, not really giving true 16 bit information. At any rate, that's a HUGE difference in tonality - probably more than most printers can deal with . . . Bob would have to weigh in on that one . . . but the point is - why limit yourself so severely when you don't have to?
Something to chew on . . . :)Cindy