Canon's new dual pixel focusing system only appears to work with a limited number of newer lenses. Indeed the way focusing points are used depends upon the lenses used.
I would suggest this means that increasingly the position of the exit pupil of the lens in relation to both the micro lenses on the sensor and also the focus sensor are becoming critical.
This then raises an interesting question in relation to testing for reviews of cameras. The trend seems to be to use a single lens and adapter to compare sensor resolving power, eg. one UK magazine uses a 100mm sigma macro lens. Now if the exit pupil is further from the sensor, that will affect the angle of light rays hitting the sensor. Now all modern cameras will be designed for their own manufacturers lenses, and increasingly also taking into account the position of the exit pupil in the design of new lenses.
We can therefore expect a camera to work best with that manufacturers modern lenses. This in turn means much of the comparison made between cameras is increasingly in error, but then this is nothing new.
Am I alone in finding magazine reviews increasingly trite. Not one review of the 7D II has mentioned to increased accuracy of its GPS due to using Russian Satellites as well as American and Japanese. One review complaining of lack of wi-fi (no mention of its security difficulties) wanted rid of GPS, yet to me GPS is the biggest boon since digital began, the end of anonymous images and hours of labelling and indexing work. The ergonomics of a camera are critical - not just is there a button to press but also how easy is it to press in use, how does the camera feel in the hand and at the eye. All subjective, but often considered by a reviewer as not right - because it is different from the brand of camera he usually uses.
I am not alone in respecting users comments after they have used a camera or lens for weeks or even months in many different circumstances.