WTF?

[AaronMC]

This a comment for this page on the website

This is so odd. It's the same sensor. The same amount of light. All that's happening is the diffraction of the AA filter is being de-diffracted. I guess that could possibly affect the ISO performance, but not DR and color.

Did you guys test multiple examples? Could this be explained by variations from sensor to sensor?

[evilthought]

AaronMC, are you sane? The difference between lowlight score is 1/15 stops. Even if they test two different D800, they will get this (and probably more than this) sample variation/margin of error.

Basically, Dxomark scores confirm there is no difference. What are you seeing in these scores?

[AaronMC]

Jeez, dude. I don't need an insult for asking a simple question.

I'm well aware that DxOMark says that small variations are within the margin of error. But they thought that it was significant enough to award the sensor another full point. I was curious as to why. If their answer is simply "it's within the margin of error and is probably measurement variation on our end," that would be a completely acceptable answer. I was curious if they thought it possible that the AA filter difference was directly responsible for the differences and, if so, how.

So thanks for being concerned about my sanity, but no, I think that I'm fine.

[evilthought]

1 point total score is irrelevant difference. 5 points are 1/3 stops, so what's 1 point difference? Nothing. It's sample variation again.

[AaronMC]

I don't think that the total score relates directly to stops of performance. For example, color depth has nothing to with exposure, but strongly affects the final score. That's why I was curious about the thoughts of the DxOMark techs.

I'm aware that the differences are academic at best, but this is the first time in digital camera history where we've had an identical sensor both with and without an AA filter. Any differences between the two are worth discussing even for just academic purposes.

[evilthought]

Read the FAQ. 5 points difference in total score on average means 1/3 stop difference. Anything lens then 5 points, the difference is lower than 1/3 stops. 1 point is irrelevant.

It's not the first time in history have you seen these differences. The Sony's 16 MP sensor is used in all these cameras (5N, K-5, K01, K30, D7000, A580, D5100) All these cameras score within 5 points of each other, meaning you won't notice any real difference in real world.