The digital images, habitually observed in a PC monitor, they are in fact a bitmap in those that each pixel you can encode by means of a group of bits of certain length. For example, a pixel can be encoded with a byte, or 8 bits, so that each pixel admits 256 variants (2 digits for bit, high to the eighth strength, that is to say, 2 x 2 x 2 x 2 x 2 x 2 x 2 x 2), for pixel. Then, in a file 8-bit jpeg, if we have 256 variants for channel and we use three channels RGB, we will obtain 16,7 million colors. This is habitually named as true color. The size in bits of the pixel is the color depth.

We will also have to know the color model that we are going to use. This way, the RGB red-green-blue) allows forming a color composing the three basic colors, where the value 0 is the absence of anyone of the three colors in the mixture and the value 255 (256 variants less 1) it means that it contributes the maximum in that color.

The color decoding formats that Nikon uses for RAW (NEF for Nikon) of their DSLR are 12 bits, what gives us the possibility to have 4096 variants for pixel or 68,7 million colors.

The most immediate conclusion that we can take out is that if we shoot in format RAW to 12 bits we will have a colors fan of 68,7 millions in front of the 16,7 jpeg 8 bits format. More color depth, therefore more quality, amen of the enormous later adjusting possibility that has a RAW file. Well, it occupies more card space, but that is another issue.

We will illustrate it. Two pictures, with flash, same conditions, same camera, one in format RAW and the other one in JPEG:

To the left, a 1600% magnification of guitar´s edge, format RAW 16 bits in Photoshop. Rightward, the same magnification but in jpeg 8 bits format. See the difference clearly in the color gradation.

ARTICLES