Some thoughts on the M series and mirrorless. Not trying to jack the thread, and it will come round tuit again - the M series that is.

HD - interesting real world handling issues that you have surfaced.

I have said a couple of times the basic functionality of the M5 on FF would be a great 6DII (mirrorless too). Same MP count, still oriented to lower light efforts. 4k would make it a gang buster for enthusiast vids.

Other than some minor expense - I guess it doesn't come up that often - why wouldn't the man'f make a 'lens change shutter' such that the sensor would not be open to the elements when changing a lens.

I am in the middle of a full on system re-eval. I get 14 bit color depth and DR - though I have never really printed to that depth - 8 bit only and my two 4k SUHD TVs only have 10 bit and that is pretty cool. I can't really understand what the leap to 14 would yield - better but how much? I have no experience for context.

As 4k and 10 bit - I watched a 4k 10bit video over the holidays - WOW! simply WOW! having said this, a few minutes into it, I was focused on the plot not the IQ. But really noticed the IQ first several minutes of the vid.

So when looking at the options for a full on system replace for the mirrorless/ or even smallish size as a 1st filter - $$ is in 2nd but as long as it is below 5k total change out costs (rules out Leica). The Pany, Oly, Sony, options are just so darn feature rich, and 99.999999% of image viewing will be on a screen converting 14 bit to 10 bit will last me quite awhile - even Dell's 8k monitor is 10bit. The unavoidable collision is the diffraction f stop that starts to degrade image quality of MF to FF due to pixel size.

So in conclusion,
I am holding on for a couple more months hoping that Canon makes an announcement that moves the "M" series into the competitive feature set. They may be aimed a bit higher on the pro/enthusiast pixel peeping, diffraction unlimited, continuum than where I think I am settling.