A little while ago I wrote a post HERE about ISO and exposure compensation. I explained that whilst I could see the practical results in the two cameras I tested (Nikon D200 and Nikon 40) and could see that the results differed between the two cameras, I did not know what the theoretical answer was to the question I asked, nor why the two cameras should behave differently.
Recently I posed the question to Ctein, and here is his answer:
On the basic level, here’s what happens:
— When you dial in exposure compensation, all you do is change the
exposure. The ‘sensitivity’ of the sensor array isn’t altered.
— when you change the ISO setting, the ‘sensitivity’ of the array is
altered by changing the amount of amplification that gets applied to
the signal that’s read out of the array. The actual physical
sensitivity of the array doesn’t change– it still collects the same
number of electrons for the same amount of incoming light.
The key thing here is that even in RAW mode, you’re never actually
looking at the charge collected by the sensor. You’re always looking
at an amplified signal derived from that charge. Exactly how this
plays out in any particular camera depends upon the camera’s
electronics and software. In effect, what kind of electronic
“characteristic curves” the designers decided to build into the
camera. That’s likely why you see different results with different