Slightly off-topic, but have you ever noticed that ever of these words carry a connotation of perversion? I.e. pixel peeper, measurebator. I think pixel peeper was coined by Ken Rockwell too?
I'm both fully concious of, and quite happy with, the fact that being a pixel peeper is a perversion
The important thing is to know you have a problem, and understand what is or isn't likely to be field relevant. A noticeable difference in the corners of a black and white test chart is often likely to be completely irrelevant in real world conditions.
At 800 ISO Left 80D Rightside 70D
OK, the problem is that they're not actually the same scene - i.e. they've been taken at different times. If you load both images into Photoshop, scale the 70D up and overlay them you can see that items have moved. There's also a different white balance on the two shots (which can affect perception of sharpness).
If you look at the black mosaic part of the label on the bottle of MUSCAT WINE VINEGAR the 80D looks quite a bit sharper. I.e. issues of depth of field and exactly where the items were placed will come into play. 80D on the left, 70D on the right:
So far, looking at the limited samples, my enthusiasm is somewhat dampened. Perhaps my expectations were too high. I was hoping that the 80D would meet the 6D, but it seems clear that the 6D still has much lower noise and maintains much more detail at higher ISO.
Either way I need another body and there's enough improvement that the 80D makes sense vs the 70D. But I'm back to also considering a 6D in spite of its limitations.
As idkdc has noted: the 6D is a full frame sensor. As such it has more than twice the surface area, so it's not really a fair comparison against any APS-C sensor.