Photographer Michael Andrew of YouTube channel Michael the Maven has put together a 'Flagship Epic Shootout Review'
video comparing the Sony a9, Nikon D5 and Canon 1DX Mark II. The full
video is genuinely worth your time, but if you don't have 44 minutes to
spend watching the full review, one section in particular is both fun
and frustrating: the blind portrait test.
Like it or not, we all have implicit biases when it comes to comparing cameras—it's hard not to when you've spent thousands (or tens of thousands) on your kit. But is the camera you say you like best, the one that produces the images you like most? When it comes down to the camera, by itself, using its own color engine, do you prefer Nikon, Canon, or Sony?
That's what Andrew wants to help you figure out, bias-free, in this blind portrait shootout. He shot 12 identical portraits using all three flagship cameras, and he challenges you to rank them before you know which camera shot which portrait. The game is simple: grab a pice of paper and list it from 1 to 12, and then draw three columns labeled A, B, and C at the top. As the images pop up on screen, give your favorite a score of 3, your second favorite a score of 2, and your least favorite a score of 1.
"At the end, we'll add the scores to discover which camera's color science you prefer most," he says. "Don't give it too much thought [...] I did my very best to take a picture of the same model, in the same lighting conditions, with the same white balance, with the same exposure settings."
It's a fun little game that can turn a bit sour at the end when you
add up the final score... especially if you've ranked your personal
favorite brand dead last. This has already happened to two of our staff
here at DPReview, and it will probably happen to a few of you as well.
And before you jump in with a "just shoot Raw" argument, our Technical
Editor Rishi has a message for you:
Like it or not, we all have implicit biases when it comes to comparing cameras—it's hard not to when you've spent thousands (or tens of thousands) on your kit. But is the camera you say you like best, the one that produces the images you like most? When it comes down to the camera, by itself, using its own color engine, do you prefer Nikon, Canon, or Sony?
That's what Andrew wants to help you figure out, bias-free, in this blind portrait shootout. He shot 12 identical portraits using all three flagship cameras, and he challenges you to rank them before you know which camera shot which portrait. The game is simple: grab a pice of paper and list it from 1 to 12, and then draw three columns labeled A, B, and C at the top. As the images pop up on screen, give your favorite a score of 3, your second favorite a score of 2, and your least favorite a score of 1.
"At the end, we'll add the scores to discover which camera's color science you prefer most," he says. "Don't give it too much thought [...] I did my very best to take a picture of the same model, in the same lighting conditions, with the same white balance, with the same exposure settings."
Which do you prefer? |
While shooting Raw helps poor white
balance issues, it’s not a panacea for a disagreeable color engine.
Putting aside for a moment the convenience of using
straight-out-of-camera JPEGs, Raw converters like Adobe Camera Raw (ACR)
include camera-specific profiles that emulate the manufacturer’s
various color modes, so if they’re not to your taste to begin with, the
Raw conversions are also unlikely to be palatable.
Furthermore, ACR can’t emulate the
multitude of non-linear, scene-dependent adjustments camera JPEG engines
perform. Even the same colors are not necessarily processed in the same
manner in a landscape as it is in a portrait. It’s hard for Raw
converters to emulate these complex adjustments unless the manufacturer
works directly with them to directly share what they’ve learned over
decades of color research.
So jump in, take the test, and let us know your scores (and whether
or not you betrayed your go-to camera brand) in the comments.
Via: Fstoppers
No comments:
Post a Comment