A tale of two cameras, part 6

Some people have reported position dependent color casts with the RX-1. I see virtually none. What’s more, I measure very little in the way of cast. Here are my results for the center and the four corners, measured in CIEL*a*b*, using the methodology described here. One unit of distance in the equation D = sqrt(L*^2 + a*^2 + b*^2), or D = sqrt(a*^2 + b*^2) for chroma errors, is usually just barely noticeable, although the eye is more sensitive to chroma errors near the gray axis.

As you can see, there is some lens falloff, but it’s not bad. There are very minor corner chroma casts wide open, but they essentially go away when you stop down a stop or more. It’s interesting that the lens falloff gets better as you stop down, as you would expect, until you get to f/8, where it starts to get worse. Maybe f/8 is where the camera software stops correcting for falloff.

The results are so good, and the camera’s auto white balance so accurate, that these are the raw numbers, with no center zeroing applied.