Why can dark-frame noise stay nearly constant between ISO 100 and 200, and why does the black level jump at higher ISO?
Asked 1/7/2019
4 views
2 answers
0
I shot dark frames on a Canon EOS 5D Mark IV with the lens cap on and identical settings except ISO. I measured the mean and standard deviation of each frame and found that noise is almost unchanged from ISO 100 to 200, then rises more noticeably from ISO 400 upward. I also noticed the black level changes from about 512 DN at ISO 100–200 to about 2048 DN at ISO 400 and above. Why would dark-frame standard deviation not scale smoothly with ISO, and what causes the black-level offset to change at higher ISO settings?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
7y ago
2 Answers
1
Based on inspection of your "captured dark frames and their associated statistics" and some experience in document scanner technology, it appears that between ISO 400 and 200, Canon shifted away 2 least significant bits, which represented most of the noise that you attempted to measure.
As for Clark's diagram, "ISOless" exposures wanting ISO above 1600 simply lack enough signal from photons to saturate the ADC's dynamic range. Meanwhile, dynamic range for exposures needing ISO less than 200 are nominally constrained by ADC bit depth. My understanding for that dynamic range bump at ISO 50 is a product decision, given that many competitive camera review image comparisons are made at ISO 100, so artificially reducing gain yields cleaner shadows and monochromatic colors.
Originally by user37166. Source · Licensed CC BY-SA 4.0
user37166
7y ago
0
Generated from our catalog & community — verify before relying on it.
This is likely due to how the camera’s readout pipeline is implemented, not a simple one-gain-per-ISO model.
From your data, a plausible explanation is that around ISO 400 Canon changes processing mode or ADC scaling. As one answer noted, between lower and higher ISO settings the camera may effectively discard or shift away the least significant bits. If those bits mostly contain read noise, then going from ISO 100 to 200 can leave the measured standard deviation nearly unchanged even though gain changed.
The jump in black level from about 512 to about 2048 DN also points to a change in internal encoding/offset, not to actual light. Cameras add a black offset so pixel values don’t clip below zero after noise and calibration. That offset can change when the camera switches gain stage, ADC scaling, or digital formatting.
So the non-linear relationship is expected because total dark-frame noise is a mix of analog read noise, ADC behavior, quantization, and downstream digital scaling. ISO does not always map to one continuous linear amplification path across the full range. Your results suggest a mode change near ISO 400 rather than a smooth progression.
Recommended products
UniqueBot
AI7y ago
Your Answer
Related Questions
Why does RAW black level stay the same with long exposure noise reduction enabled?
Why does my webcam produce a perfectly black dark frame with no visible noise?
Why does my camera estimate fewer RAW shots fit on a card at higher ISO?
How does film ISO affect black-and-white development time?
Is ISO 200 a sensible choice on modern digital cameras?