He does refer to the ‘real sensitivity’ of a sensor which may be the same thing I’m saying, but I’m an empiricist where he is an engineer.

Simply stated, I know it comes out better in high contrast situations when I underexpose one stop in the 5D then use Aperture to restore the shadows. Try it and see.

Another alternative with high contrast scenes is multiple image HDR but, let’s face it, it’s a real pain to do, and limited to stationary subjects.

I hope having inadvertently reminded you of something annoying, I’ve provided some reasonable reading compensation

I don’t know and, frankly, I no longer read LL owing to their dishonest reporting on the Leica M8. When I want fiction I don’t go the the fact section. How can you ever trust this fellow ever again when he as much as admits to being in the pockets of the makers?

However, it’s probably a combination of the fact that the 5D’s sensor is 1/3 to 1/2 stop more sensitive than indicated (maybe the exposure determination discounts this?) and older sensor design. It’s interesting to note that the specs of the yet to be released Canon 1D Mark III provide for dynamic range extension with the push of a button – they illustrate it at work with a bride’s wedding dress. Dynamic range extension has to be the next step in sensor design (I hope) – meanwhile, I stick with underexposing on contrasty subjects.

We will see.

Is it that Canon has already pushed too far to the right on the 5D and so we need to underexpose?

Thanks in advance!