OpenEXR folks just published a general article about scene vs display referred processing, could be a nice introduction for beginners:
Thanks for sharing !
Allow myself to also add this recent video:
Additive mixture
Daniele Siragusano on “scene-referred”, image states and maintaining ratios.
Which also tackles the scene-referred concept and allow to explore the concept further.
OMG… someone eventually explained it in terms that makes sense to the average person like me.
“Scene-referred” means the pixel values indicate how much light was received by the camera. “Linear-light” means the numbers in the image are directly proportional to the amount of light they represent. Most image formats are display-referred, where the values indicate how much light should be used to display the image on a screen…"
The manual actually does a decent job I think and frames it in the context of how things work in DT… especially noting how modules with 0-100% UI restrictions behave…
https://darktable-org.github.io/dtdocs/en/special-topics/color-pipeline/
These couple of sentences just explained it so succinctly. I had previously listened to AP’s description on his videos and I just find this so much easier to understand.