There's a new review of the Nikon D750 at DxO and in it there is this.
'For low-light ISO performance the D750 certainly holds its own too, with a score of ISO 2956 ranking it in 8th place overall. This is a very similar result to ISO on Nikon’s D600, D800E and D4 sensors and we can essentially say they offer the same quality in low light, despite the expanded sensitivity up to ISO 51,200. As you’d expect however lower resolution full frame sensors, such as those found in the Nikon Df and D3s, do offer a modest improvement over the D750 for ISO.'
So the Nikon D800E has the same low-light high ISO performance as the Nikon D4 does it? OH NO IT DOESN'T. Here is their D800E and Df comparison.
Here is the selection of lenses they used on the above cameras to come to their conclusion.
Interesting !!!!!!! selection. All wide-angles, and wide wide angles at that. Strange choices.
Now the statement above isn't even matched by their own figures. And having owned and used a Nikon D800E extensively I can assure you that it isn't actually very good at all at ISO's over 1600. Further to that the notion above that the D800E has better dynamic range than the Df is also total BS.
Talking of low light performance, lets see how DxO rate the Fuji X sensor. No wait, we can't. DxO haven't tested any Fuji X sensor.
Oh and they haven't tested a Sigma Foveon sensor either.
So it seems they can only 'test' Bayer sensors. So what use is this? you might ask. And the answer seems to be not a lot. Their lens tests are OK but then there are lots of sites that can test lenses.
As far as their sensor tests are concerned I give them no credence whatsoever. Their results just don't bear scrutiny. If they can't test a Fuji X-Trans sensor then they have a fundamentally flawed testing procedure and one that is obviously not capable of testing non-bayer sensors. They do like a Nikon though.