Pixel fetish

[Rick Burnett]

I agree more DR is a very desirable characteristic for sure, but in view of emerging production markets the balance between resolution and DR needs to be carefully considered and one does not necessarily have to trade off resolution in order to increase DR, nor is decreasing resolution likely to yield huge gains in DR for a given size sensor.

Agreed. Basically, it's not a one for the other in simple terms, it is MUCH more complicated than that and from your post processing example it is very very true. Same with the active pixels that contain multiple active devices, this is essentially enhancing the resolution, although just duplicating the same bits, but so that you can try and remove some of the noise.

After thinking about this more, I think the real improvement to CMOS design in general would be in sensitivity. If you held noise standard between a ISO320 equivalent sensor and an ISO800 in terms of photon detection only, for the same exposure time, the ISO800 would accumulate a much larger value more quickly. With this, you could store that intensity level in a larger capacitor which could yield even greater dynamic range since the noise floor would be the same. Of course, a deeper bit-depth A/D will probably add more noise as it would be bigger.

Which goes to show, when designing these sensors, there are about 23094023 variables to consider

 

[David Rasberry]

Compare today's cameras with my 4 year old Canon GL2 DV camcorder which has a practical ISO range of about 50-200 with the sweet spot for maximum DR of about 7 stops at 100 ISO (figures from my own tests and shooting full manual exposure control film style with a light meter).