For deep sky EEA the 3 main priorities are :
Sampling is only a secondary problem as the goal is to display in few seconds an
image of the target. Undersampling is not a problem. And it's a way to increase
incoming light on each pixels.
The bigger pixels are, the better (of course read noise has to be lower as possible,
it's a fondamental parameter). A camera with big pixels coupled to a fast telescope
is the best solution for deep sky EEA.
That's why IMX 482 is clearly the best new sensor for this use. And a priority for
a new color camera that will be the best for EEA, short exposure imaging on faint
objects and at the same time a good planetary choice (90 fps is not the best frame
rate available but it's enough for serious lunar/planetary imaging)