Quote Originally Posted by Colin500
And it's "unfair" to compress 11.5 stops of dynamic range into an 8bit jpeg, and/or render it to an 8bit display, because it exaggerates the noise?

The noise is the fault of the camera, it is not jpeg artifact, nor is it caused by the display (ie, 8 bits isn't the problem).


But yes, it is "unfair" because the shadows have been brightened far more than a "standard" jpeg converter would have done, which allowed us to see shadow detail but also exaggerated the noise.


When I first saw Daniel's post, I thought it was misleading. I saw the jpeg and thought, "Yeah, but any camera will show noise in the shadows if you brighten them enough. In order to know how bad the read noise is, we need to know how much the shadows were brightened."


When I re-read the post, however, I realized that the point of the jpeg was to demonstrate what read noise looks like, not to evaluate how much of it the 5DII has. (And of course, Daniel later told us exactly how much the shadows were brightened).