@David.

There was a bit more difference back then, especially since sensor performance was not nearly as good as it is today. So those differences definitely were important back then.

But looking at the down sized R5 results. They are within a very small margin of each other indicating very minimumal loses. I know others will point out read noise, but without looking at a dark image downsized to the same dimensions. I'm going to bet money they will more or less parallel the well lit ones. Simply because they are the same basic sensors with the same tech for both. And because of how over all noise is only fairly judged at the same output. Not at the pixel level or the read out noise at the pixel level.

Now there is a place for knowing how good the noise/overall image quality is at a pixel level. Very large enlargements is one such application that comes to mind, you want awesome pixel level image quality at it's max resolution.

I think my argument is that some are probably not using a more suitable camera because they believe it's "Not useful" to them because they just look at 100% crops and are misled thinking it's just magnifying issues when it's actually not. When a high resolution body like the R5 or 5Ds R is actually what you really want. Think of it as lower resolution adding another level of IQ disadvantage and blurriness (Think big enlargements here) vs have more than double the resolution and defining the scene much better. Sure it will mask blemishes in a large print, but in the worst way possible. By bluring it. It's not what you want.

Now this is not an argument for those that simply have no need for high resolution or those that are totally fine with say 20mp and rather have say the benefits of smaller file size, cheaper bodies, cheaper computer to run effectively ect. Those are valid points.

I'm just saying the way many look at high resolution, especially how it pertains to noise and defects in the image, is not correct.