What’s all the fuss and hype about RAW files? Let’s look at a little story as a comparison.
The Emperor’s New Clothes
The Hans Christian Andersen story of an incredibly vain King is an amusing tale with an interesting moral.
One day the king, who was very fond of fine clothing, was approached by two slick-talking swindlers. They posed as weavers, and they said they could weave the most magnificent fabrics imaginable. Not only were their fabrics uncommonly fine, but clothes made of this fabric were invisible to anyone who was unfit for office, or who was unusually stupid.
“Those would be just the clothes for me”, thought the Emperor. “If I wore them I would be able to discover which men in my empire were unfit for their posts. And I could tell the wise men from the fools.” As the story goes, the king bought into the story and the clothes. As a result, the people of the kingdom discovered more about their king than they ever cared to know.
The truth occasionally gets lost in marketing hype, even in photography. How many times have you heard the claim that vast amount of visual information can be seen in RAW image files? There’s a major problem with that claim, the same problem that “exposed” the king in all his vanity. The claim ain’t exactly accurate.
RAW files do indeed contain all the information collected by a digital camera’s image sensor. But the file’s information itself cannot be viewed because the RAW data is not an image at all, it’s merely numbers.
Only when these numbers are parsed (interpreted) as colors and tones by special software can they display any visual information. RAW Interpreter software builds an initial visual image from the data in the file.
The RAW image, just like the ill-informed Emperor’s clothes, doesn’t actually exist until the file data is interpreted. There is no such thing as a RAW image, only RAW data.
When you do open a RAW file in Camera Raw, Lightroom, ON1 Photo Raw, Alien Skin Exposure 3, etc., the image you initially see on the screen is actually based on the camera’s built-in JPEG expression of the RAW data; a mere rough draft of the file’s potential information. The camera’s exposure settings (recorded along with the RAW image) determine the file’s initial appearance on the computer monitor.
Once this initial image appears on the monitor, each RAW Interpreter software provides a fairly exhaustive array of color and tonal sliders that can shape the data into a variety of interpretations. Each interpretation can be saved in JPEG format and published for others to see. Folks who shoot and publish JPEGs directly out of the camera are really shortchanging the file’s potential and leaving important color and detail on the cutting room floor.
The RAW Truth
The term RAW is not an acronym for some technical phrase nor is it a reference to some uncooked food. It is merely a coined word describing the collection of undeveloped (latent) image data from the camera’s image sensor. This data file contains all the raw chroma and luminous data extracted from millions of light buckets called image receptors located on the camera’s image sensor. Each light bucket is covered by a blue, green, or red filter.
These RGB filters split the incoming light into three channels of information. Each receptor records the strength of the filtered light as an individual color that will eventually form a single pixel in the image.
While the initial grid of receptors is covered with more green filtered buckets than red or blue, the purpose for this imbalance is a bit too complicated for this article. Suffice to say, the image processor in the camera performs some very complicated math to determine each pixel’s color value and brightness.
This light capture process begins even before the display is visible on the back of the camera. Every time you reposition the camera to frame your shot, the image processor does its magic again and delivers a new preview of the composition. If your camera is set to display a pre-capture histogram of the scene, this processor data is used to simulate the graph on the histogram.
But the real heavy-lifting happens when you push the shutter button and the image is captured. Once all the individual colors are recorded on the sensor and delivered to the processor, the final image information is preserved on the camera’s hard drive.
In a RAW file, the value of each pixel can be extensively adjusted for hue (color), saturation (intensity), and luminance (brightness). JPEG files record pixels with the same initial color values but the JPEG file format significantly restricts the ability to adjust those values in the editing process. The latitude of JPEG adjustments is significantly limited.
JPEG files record each color pixel as an initial luminance (brightness level) and chroma (color) value. When all the pixels on the grid (bitmap) are collectively interpreted in imaging software, a visible replica of the original scene appears on the monitor. If that same image is also captured as RAW information, the values of luminance and chroma are captured in the context of a larger color space and can be interpreted in a wide variety of expressions of the original scene.
RAW files have been likened to photographic color film negatives in that when they are “developed” (viewed in RAW Interpreter software), the image can be “printed” (published) in a number of unique colors and tonal versions.
But the truth is that because this RAW file is not an image per se, but a record of the light characteristics captured by each of the camera’s light buckets, the original image data contained in the RAW file never gets altered, it only gets interpreted.
The interpretations are records of the luminous and chroma adjustments made to the RAW bitmap pixels. These interpretations are what gets saved as JPEG images.
Unlike the yarn spun by the king’s “couturiers,” RAW data files deliver custom-tailored results and can make you look really smart in a couple of ways. Dress your images for success.