mpmendenhall Offline Upload & Sell: Off
|
Bifurcator wrote:
Hey, that's an interesting test. Have you tried this with other cameras? What were the results with them?
Here's the same test with the one other digital camera I have to compare, a 5Dc, using a C/Y 100/2 at ISO100, f5.6. I've tried to keep the processing identical on the "boosted" version as for the DP2M, but to do a quantitative comparison you should look compare the unmodified shots:
http://praetoriusphoto.images.s3.amazonaws.com/fmforums/20121011_dp2m/greycard_5d.jpg
http://praetoriusphoto.images.s3.amazonaws.com/fmforums/20121011_dp2m/greycard_5d_sat.jpg
The 5D sensor color shifts have a different characteristic, following the sensor edges and side-to-side variation rather than a radial dependence (so color shift in this case is not correlated with light angle from the lens, though this could also be lens dependent). Color "blotching" is present with a finer-grained structure. The overall color variations, based on looking at the a,b channels in Lab color space, appears to be roughly half that of the DP2M; perceptually, the 5Dc gray card shot looks a bit more uniform and neutral to me.
Perhaps people with access to other camera bodies can post their results here? I'd be interested in seeing how newer sensor tech than the venerable 5Dc behaves (Canon 5DII/III and current Sony/Nikon sensors).
|