Do you have any suggestions for a fairly simple dynamic range comparison that I could do? I was thinking of perhaps taking an incident meter reading, set both cameras to that setting, and think bracket + - 1-stop, and then messing with each of the camera's 3 files, to get the most detail, then comparing the best result from each camera. Geez...even that seems like a lot of work. ;-)
Oh, I can think of many. ;-) But I think I would go about it as follows: I'd first try to calibrate the ISO settings of the two cameras by shooting a white wall or the like at the same exposure. You would start at the same ISO on both cameras and then change it on one of them until you reach the same level of brightness. Since difference in lens transmission values (f-stops are nominal too) is another potential source of error, you should use the same lens on both cameras (I trust you have an m43 adapter for your Nikon lenses), or, even better, two different lenses which you already know to have identical transmission, or which you can calibrate in the same way as with the ISOs, and which, in addition will give you about the same FOVs on the two cameras (which simplifies things in the field later on). Of course you should choose an aperture where vignetting is not a problem.
Once you are done calibrating, you shoot a contrasty scene with both cameras at the same
real ISO (somewhere around the 400-800 mark), and the same
real exposure (taking into account any difference in lens transmission values). Underexpose so as to save the highlights. You may want to try two or three different levels of underexposure to have enough to play with later on.
Then choose the pair of shots that looks most representative of what you ordinarily get/do and PP them in the same way. If you have done your calibration work right, this should suffice to reach identical brightness in the shadows as well as the highlights. Finally look at the shadow noise to see how much of a difference there is. Make sure you do this at identical display size, rather than at 100 percent for both cameras since the latter method will bias the results against the GH2.
One way of simplifying the whole thing is to simply shoot a shadowy scene at the same
real ISO and the same
real exposure and just look at the noise level. It's that level which will ultimately determine how much you can lift the shadows.
Well, AF accuracy is certainly one area where I am less than perfectly happy with my Pentax. And judging by the results provided by optycne.pl, which is the only review site I know of that actually tests AF accuracy, this is a problem common to all but the very best, biggest, heaviest and most expensive DSLRs. As expected, the D300 does better in this regard than many others, but as you can see here (under the headline "Autofocus"), it still cannot match the GH2 with respect to accuracy:
http://translate.google.se/translate?hl=sv&sl=pl&tl=en&u=http%3A%2F%2Fwww.optyczne.pl%2F60.3-Test_aparatu-Nikon_D300_U%25C5%25BCytkowanie.html
http://translate.google.se/translate?u=http%3A%2F%2Fwww.optyczne.pl%2F149.3-Test_aparatu-Panasonic_Lumix_DMC-GH2_U%C5%BCytkowanie_i_ergonomia.html&sl=pl&tl=en&hl=&ie=UTF-8
Interesting. What are they graphing in those AF performance charts? I don't know what is on the X and Y axes.
The charts show the frequency distribution across 40 test shots. The leftmost bar shows the number of shots that are essentially perfect with regard to accuracy (less than 1.5% percent deviation from that shot in the series which shows the highest resolution). Bars further to the right show the number of shots that deviate from the ideal. The further to the right the bar is, the greater the deviation from the ideal.
As you can see, 35 out of 40 shots (87.5 %) end up in the leftmost bar for the GH2. For the D300, the corresponding result is 7 out of 40 (23.3 %). Still, the D300 does pretty well compared to most DSLRs I have looked at since most of the remaining shots are in the first two bars to the right of the leftmost one. Many cameras show a considerably greater spread.
