Eki Halkka
Well-known member
- Joined
- Sep 24, 2008
- Messages
- 515
- Reaction score
- 0
- Points
- 0
- Age
- 57
- Location
- Helsinki, Finland
- Website
- www.kolumbus.fi
Just look at the chart at the beginning of this thread. The Red low end is much less blocky and more smooth. The Alexa runs into the whites better.
Rolling of the highlights is a total non-issue: both cameras have linear sensors, as far as i know, so half of the values the cameras gather are in the highest stop. That means, Red has 2048 brightness values between the two brightest wedges in that chart, assuming the brightest wedge is exactly at the brink of overexposure (which it seems to be: examining the jpegs, some pixels are clipped 255,255,255, but some pixels have some variation, so it must be almost exactly at the brink).
Same goes to Alexa - the brightest wedge is at 240, and there's very little variation. I assume that it too is (correctly) exposed to the brink of overexposure, and the lower white level is simply the result of conversion to the jpeg we saw.
In other words, there's always more than plenty of information in the highlights, unless they are clipped.
How the transition to whites is performed, is a color correction issue, not a camera issue. What we are interested in is the shadows: how much we can underexpose to protect the highlights.