Higher bit depth in a raw file does not directly prevent pattern noise but makes its effects less visible. A raw file with a higher bit depth provides more tonal information, which improves the camera's ability to distinguish fine details from subtle noise, especially in underexposed areas.
Bit depth and signal resolution
Bit depth is the number of bits used to describe the brightness and color of each pixel.
- 16-bit raw files, common in high-end systems, offer 65,536 tones per color channel.
- 12- or 14-bit raw files, more common in consumer and prosumer cameras, offer 4,096 or 16,384 tones per channel, respectively.
- 8-bit JPEGs compress this to just 256 tones per channel, which can lead to visible banding in areas with smooth tonal transitions.
How bit depth reduces visible pattern noise
A raw file with higher bit depth is more robust against the visual impact of pattern noise, which is fixed-pattern noise from the sensor's electronics. The following factors explain this effect:
- Minimizes quantization error: The analog-to-digital converter (ADC) in a camera rounds the analog signal from the sensor to a digital value. This rounding, known as quantization, introduces a tiny error. A higher bit depth provides more "steps" during quantization, making the error significantly smaller and less noticeable.
- Lowers the visible noise floor: The "noise floor" is the level of inherent electronic noise in the signal. Higher bit depth increases the dynamic range and lowers the noise floor relative to the total signal. This extra dynamic range, which is approximately 6 dB for every extra bit, provides more headroom above the noise.
- Improves shadow detail: The effect of pattern noise is most obvious when you "lift" the shadows in editing. A raw file with a higher bit depth has more subtle tonal graduations available in the darkest areas. This provides more leeway to brighten shadows without revealing the stepping (banding) or introducing color shifts that would occur with a lower bit depth.
- Distributes noise over more values: In a low-bit-depth file, the limited number of tonal values can force noise patterns to be compressed into a small range, making them more visible. With a higher bit depth, the same amount of electronic noise is distributed over a wider range of possible values, making the subtle patterns less perceptible.
Important caveats
Higher bit depth is not a guarantee of low noise: While a higher bit depth offers benefits, it does not erase the underlying physical noise of the sensor itself. An underexposed photo from a 16-bit camera can still have more noise than a properly exposed photo from a 14-bit camera.
The benefit is most apparent in post-processing: For a perfectly exposed image, the difference in noise visibility between 12- and 14-bit raw files may not be noticeable. The key advantage of higher bit depth becomes clear when pushing exposure and tonal adjustments in post-production.
