Count me in.Ok: you didn't understand what I said.
My understanding how a pixel well in a sensor works is as follows:Which is that no sensor has a single curve - you have a range of curves for (at least) different contrast levels. A lot of consumer level sensors over specialize in the high contrast part of the curve at the cost of the lower contrast level. This is why pro cinema cameras and lenses often very different characteristics to consumer stills gear, with eg very low pixel counts relative to sensor size and thin filter stacks. Eg an Alexa 65 has 20 times the sensor area of an RX100 but the same pixel count.
- with no light there are a few noise photons inside.
- with increasing light the well is filled in a quite linear relation.
- when the capacity of the well is reached more light does not increase the number of photons captured which is called clipping
My understanding until now was that all sensors are (more or less) linear and the curve is implemented in software in the JPG engine.
Last edited: