> How would you suggest testing the subtle color response between the
> A7 iii, A7R iii, and A7r iv?
That's really an age-old problem. Typical camera tests involve a
standardized setup with calibrated targets and miscellaneous objects.
I've even built my own setup that I've found to be quite helpful.
However, the problem is that ALL cameras are designed to pretty much
match each other under the standardized setup that is aimed to
identify camera performance at the sweet spot. We can, and do,
extrapolate with additional +/- tests to see what the DR and noise
characteristics are like.
But what is missing is the usability testing which is unique to the
bias of each individual. For example, I tend to hyper-correct my
images to an artistic goal. But not always. When shooting portraits, I
want that flash-lit photograph to be dead-on, straight-out-of-camera
with zero adjustments. With my cameras, the E-1, L1, and GX85 are
exceptional in this regard. The 6D and E-3 required significant
adjustment. However, when shooting the Great Alaskan Landscape,
absolutely no camera comes close to what I want without major
adjustments. Of the Kodak Sensor cameras, I think the E-1 comes the
closest, but it might be my mood because the E-300 might be a little
better. To achieve my artistic goals with the landscape shots, I've
get extremely aggressive with the settings. If some of those Lightroom
sliders that have a +100/-100 range were to be expanded to +200/-200,
I'd be happier.
And that's where the subtle color response comes in. When all the
settings are left at the middle default position, we generally see no
significant variation because the imaging chain is optimized to this
standardized point. However, if I have the -100 Highlight, +100
Shadow, +50 White, -50 Black, Texture, Clarity, Dehaze, and Vibrance
set to +25 there is a massive amount of bit-bending going on. With
this amount of bit-bending, those tiny, subtle differences become
highly exaggerated.
>From what I've seen, I think the A7R III is one of the best cameras
around when the pixels get punished like this. The Canon files just
don't tolerate it nearly as much. The GX85 files are definitely not as
tolerant of this abuse for color, but are really good for B&W extreme
actions.
The one question you are dying to ask is "how tolerant are the
E-1/300/400 files to this abuse?" Honestly, they don't tolerate
highlight recovery well at all. Highlights will go funky in a
heartbeat. Otherwise, the files are forgiving. What is most
surprising, though, is how much less adjustment is REQUIRED with the
Kodak sensor files to achieve the same gawd-awful levels of Disney
which is my style these days. It takes a massive amount of adjustment
to the CMOS files just to bring them to a slight adjustment with the
CCD files. A +50 adjustment with a CMOS file may only need +10 with
the CCD file. And the less forced bit-bending we have to do, the more
color and tonal integrity we get in the output file.
So to answer your question, I'd take a standardized shot with the
cameras to test, and then jam the sliders hard left and right (every
single one) in Lightroom to see how the files respond.
AG Schnozz
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|