I like the way that Bill Claff presents similar data (in his case, dynamic range to resolution-adjusted target SNR) better:
http://www.photonstophotos.net/Charts/PDR.htm#Nikon%20D810,Nikon%20D850
For one thing, Bill plots the results in 1/3 stop increments, as opposed to DXOMark's full stops. That allows you to, in the case of the D850, see the result of the conversion gain change, and to see that you shouldn't use ISO 320 if you care about DR.
DXOMark does have the advantage that they plot actual sensitivity versus nominal, though.
I've alwaysbeen wary of using intermediate ISO settings. They didn't seem to serve much purpose years ago when i was using Canon DSLRs. It was generally better to underexpose at ISO 200, for example, than attempt to get an ETTR exposure at ISO 320 using the same shutter speed. Shadow detail (when shooting RAW) would tend to remain the same at ISO 200, and the risk of blowing highlights would be reduced, compared with the ISO 320 setting.
My take is that, between the D810 and the D850, PDR is essentially a wash until ISO 400, where the conversion gain increases.
In practice, for general, hand-held photography, when getting a sufficiently fast shutter speed to freeze movement is often a concern, I find that ISO 200 is a much-used setting.
According to the DXO graph, both the D810 and the D850 at the nominal ISO of 200 have exactly the same measured ISO of 141,
yet the DR of the D850, at ISO 141, is half a stop better (or 0.49 EV to be precise).I would consider a half stop of greater DR to be at the threshold of a worthwhile improvement, although less than half a stop, such as 1/3rd or 1/4th EV, would be of little concern. Wouldn't you agree?
I tend to value DXOMark's DR results because I have always found them to be in approximate agreement with my own practical tests comparing shadow detail and noise in identical scenes shot with different cameras.
