You are not going to be fully satisfied as the interior is not as contrasty as I had first thought. These two images are the closest I can find to the exact same settings. I see that the S2 is slightly sharper and the there is slightly more definition in the black of the piano cover. What's interesting to me though is that the Nikon needed to be exposed 1.5 stops more to get the same visual density, even though the S2 has only a 2/3 stop advantage on the base ISO. All LR3 settings are equalised between the two images, sharpening and NR is set to zero.
Update : I downloaded and looked at the file as you would see it and there is some jpeg artifacting in the piano cover black - this does not show on the original. The rest of the images show the relative differences reasonably well.
Nick,
Thanks for taking the trouble to post these examples. It certainly
appears that the S2 is the equal of the D3X regarding dynamic range. At normalised print sizes of full scenes with the same FoV, the S2 should therefore have slightly
better DR than the D3X.
It will be interesting to see what the DXOMark results for the S2 will be. For all we know, its DR might be the equal of the D7000. Wouldn't
that be remarkable!
However, from my perspective as the viewer, there are too many uncertainties in this comparison for the purposes of assessing DR. The highlights are clearly blown and there appear to be no deep shadows containing detail, only dark cloth with a low reflectance.
As I mentioned earlier, at a signal input of 1% grey (on the log scale) the Canon 5D2 has an SNR equal to the P65+, at the pixel level, according to DXOMark. It's only
below that input level that the P65+ edges ahead of the 5D2.
The Nikon shot might be overexposed, for all the viewer knows, which creates another uncertainty. By the way, I don't see 2.5 secs exposure, as opposed to 1 sec exposure, being quite 1.5 stops greater exposure. 1.5 stops would be 3 secs for the Nikon. It's more like 1 & 1/3rd stops more, isn't it?
DXO rate the sensitivity of the D3X at ISO 78. Double that and it's very close to ISO 160. The Leica M8 has a DXO-rated sensitivity of ISO 151 at the nominated ISO 160, so it's quite conceivable that Leica in this case are spot on with their ISO 160 rating for the S2.
The additional 1/3rd of a stop exposure you gave to the D3X could be explained by the different T-Stops for the respective lenses used. Again, DXO come to the rescue. They've tested the Nikkor 24-70/2.8 and their results indicate the T-Stop at 70mm and F2.8 is approximately F3.2.
I can find no T-Stop results for the Summarit-S 70mm, but it might be reasonable to presume, because it's a prime lens and a very expensive lens, that the T-Stop is the same as the F/stop.
I'm not sure if one can presume that the T-stop at F11 (for the Nikkor zoom) will be F12 or F13 if it's measured as F3.2 at F2.8. Perhaps someone can advise me on that.
Merry Christmas!