Understanding Raw

People often misuse or misunderstand the term Raw. In the context of digital cinema cameras, Raw refers to capturing the raw voltage information from a camera sensor. Raw does not mean uncompressed nor it is not the correct term to generally describe any camera's 'source' footage.

What gives? RAW sounds like it should be Uncompressed

Raw can be uncompressed but it does not necessarily have to be. Creating a viewable image is a complex process. After the camera collects the initial data from the sensor, the image requires additional processing. Traditional digital cameras perform this processing in camera, locking you into certain decisions about what your image should look like such as white balance and ISO.

To understand Raw we have to take a quick lesson in imaging technology. Modern digital cinema cameras use a single chip CMOS sensor. Believe it or not, that sensor cannot inherently see in color. When a photon (particle of light) hits the sensor, it simply records the voltage information at that particular point (known as a photosite). These voltage measurements are taken for every frame across all photosites on the sensor. The number of photosites on the sensor generally defines the camera's resolution and the number of frames per second defines the shooting frame rate.

So how do you make a viewable image out of voltage measurements?

Each photosite has a red, green, or blue color filter placed in front of it. The most common arrangement of the color filters is known as a Bayer pattern. A Bayer pattern is a specific alternating pattern of red, green, and blue color filters, where green is the dominant filter. The RGB color system (which is used for displaying color on computer monitors) is an additive color system which means to derive the full gamut (range of color a camera is capable of capturing) we need to know the Red, Green, and Blue information for every photosite. While some camera systems dedicate a sensor for each of the three color channels, most modern digital cinema cameras only have a single sensor and we have to apply some fancy math to calculate the values of the two missing color channels for every photosite. This process is known as debayering. For example, a red photosite looks at its neighbors to try to figure out what the missing blue and green components are.

Ok so now tell me what RAW is again

RAW is the sensor data before it has been debayered (converted into a 3 channel viewable image). Sounds simple now, right?

Ok so now explain the difference between RAW and Compression

Compression typically refers to throwing away some portion of your image data to make files smaller and easier to transmit. Raw images can be compressed just like RGB images. In fact, compressing a Raw image is significantly more efficient than compressing an RGB image because there is only a single channel black and white image.

So what's the benefit of shooting RAW? Sounds like a lot of extra work!

Image quality is the main benefit of the Raw Digital Cinema workflow. Recording Raw allows you to capture the full dynamic range that the camera is capable of recording into a reasonably small package. Shooting Raw also allows for key creative and technical decisions to be made in the calm environment of a post studio. During a grading session at Katabatic, the colorist has full control over white balance, ISO, color science, and other key processing controls. Without the Raw workflow you would otherwise have to make these decisions on set. In Katabatic's color grading suite, a Raw workflow means major production mistakes can often be fixed in less time than it takes for you to sip an espresso, creative processing choices are endless, and you can be confident you are producing the highest quality image possible.