"You might find it interesting to photograph a ruler with well contrasting ruler marks, at about a 45 degree angle, at a very close distance at or near minimum focus. Not to test the point of focus, but to test the visually perceived DOF in the viewfinder by estimating the range of ruler markings you can see with your eye and then comparing to an actual image. Do that at 1/3 stop aperture increments from wide open to about f/4. I suspect you will find that they correlate only at about f/2.5 or smaller."
I think the D600 viewfinder is measured at f/2.2, I maybe even read that in the manual. Certainly with the D300/200 using apertures of less than f/2 the DOF in the viewfinder was noticeably greater than in the final image, which felt line witchcraft - I recall that it all equalised at about f/2.8. I noticed this when shooting the 85mm f/1.4 wide open - it was like getting unexpected extra boke... (it's in that old thread in the lens forum to which I yesterday added a post)
"I suspect that camera could easily go to 100K but Nikon basically feature crippled it? Probably high enough for most work and the camera might do ok with an intentional 1 stop underexposure to simulate ISO 50K?"
I had already got that one planned - it's pure marketing I guess - the D600 beats the D3 in low light, and the D3 went to over ISO 100k.
Of course this does beg the question - one would presume that getting this sort of performance in a DX camera with a 10.5MP sensor would not take much at all. So there's your D400 - perhaps squeeze it to 12 MP, set the maximum ISO at 25,600 and push the frame rate to 10 fps? I have a feeling that it won't happen though. Given the first two entry DX cameras are now knocking out at 24MP and have the same maximum ISO of 6,400 as the D600, I am not sure Nikon would consider that the mass market would go for such a camera, and if they did it might hit D4 sales?