>>3363970>Please, tell me how pmp is nonsense, Megapixels is a measure of SENSOR SAMPLING FREQUENCY and is NOT a measure of RESOLVED TARGET DETAIL. There is no standard definition of how a 10 or 20 or 100mp image should look therefore there's no reference for saying a sensor + lens is perceived as 10 or 20 or 100mp.
It would be like trying to compare cars by "perceived RPM." It's utterly meaningless.
I've compared two sensor/lens combos where they reported one as having half the "perceived megapixels" of the other. The images were in fact very close, much closer than two images using the same lens on a camera with 10mp and 20mp.
>i can't help but notice you didn't expand on that at all, and as dxo offers all their method and evidence the burden of proof that it's "wrong" is on you.If dxo says A is better than B and RAW files from other sources like dpreview and IR look identical, or show B to be better, they've been proven wrong. The graph or score HAS to match reality, not the other way around.
>Color bit depth is far from meaningless, It's utterly meaningless. Take the lowest color depth ILC sensor on the chart and the highest color bit depth sensor. Now take the same photograph with both and display on a 4k IPS display or print to a pro ink jet. No human being will be able to see a difference in tonal/color gradation between them. For all we know the color bit depth test is accurate, but it represents differences below the perceptive threshold of human observes so who gives a fuck?
>Again, you claim the high iso tests are observably wrong, but left no observations.Pic related. dxo ranks the cameras this way: A73 > A7S > A7r3 > 5DsR
My eyes see: A7r3 > 5DsR > A73 > A7S
dxo probably gets high ISO wrong because they completely ignore resolution and treat everything as if it were 8mp.
