Hi,
funny you asked ;-) I just did a D50/D70-comparision today @ 1600 ASA in raw. Well, it's very close. The files do look a little different out of the camera, but if you apply the same settings to both in the raw-converter they come very close. If you do a little tweaking they come even closer. The difference is very subtle. I think, the D50 provides somewhat brighter and more contrasty files than the D70. Also the noise of the D50 looks a bit more diffused (unsharp) than from the D70. The shadows in the D50-NEF were darker than in the D70 NEF. So there was more noise in the D70 NEF, but also more shadow-detail. When I turned the D70-NEF to darker shadows and brighter lights with a little adjustments at the colors both files looked very much alike. In the end the D70-NEF had somewhat crisper noise than the D50 NEF and showed slightly but visibly more detail. The latter might be due to minor focusing differences, but I preferred the noise of the edited file from the D70.
So for me there is no difference in the data from the sensor, it's a matter of editing. The differences between severeal raw-converters arw bigger than the differences of raw-files of the two cameras in question. I used capture 4.4.2.
Some time ago I tried D2x/D200/D70/D50. At 1600 ASA the D70 had more detail and less noise than the D2x. At 800 ASA the D2x took the lead. The D200 was very close to the D70, it was a tie. The D50 looked better and sharper than the rest (all at 1600 ASA). But at that time I didn't edit the files and compared out-of-camera-NEF. Sooo, if the D80 is very close to the D200, then at high ISo there isn't much reason to change to another nikon IQ-wise. Of course, the newer ones are faster to operate.
HTH
Greets
Ralf C.
