So my point was that the blue dots represent the peak pixel value, these you can see start to saturate and longer exposures don’t increase the peak value linearly – the orange dots are the sum of all the values in the central apature, when you sum these up, the result appears to continue to be linear.
My point really is that whilst the advice is ‘stay within the linear range of the sensor‘ – which I do do – for my sensor at least – the actual practical process of counting the electrons is done not on single pixels but on a defined area of pixels and when this is taken into account the linear range seems to extend well beyond that of the linear range of individual pixels, and if this is the case it might have some practical application.
BTW I’m not really disagreeing just pointing out that it seems interesting… 🙂