My hand is directly linked to my arm, but it's not the same as my arm.
Using the definition of DR as the ratio between the lowest amount of light captured, what kind of link can that be? With all due respect, that doesn't make sense to me. You can have great DR with little to no colour depth. 2 bit colour can have as much DR as 14 bit colour. All that has to happen is that the white (11) that is captured is 15 EV brighter than the black (00).
But even if it's true, is there enough difference between 12 bit raw and 14 bit raw to be of any importance?
I've investigated this, and found no discernible practical difference between my 14 bit K20D and my 12 bit K-x.
I've posted examples, and explained why I think the way I do.
Please take some example shots and show me why you think ETTR is better, or demonstrate that more colour depth leads to more DR?
What I'm getting back is theoretical, and what I consider to be questionable DxO analysis in which they co-opt common photographic terms and misrepresent them.
To me this is focussing on little bits of technical trivia of no practical use. As with most things DxO, you know what they said, but what they said is disputable and controversial. And as far as I can tell of no practical value to photographers.
Please show me why the way you think is more appropriate, with images? After all, it really doesn't matter if there's a difference between cameras powered by hamsters on wheels and cameras that use mice, if it makes no difference to my images