This is a fun little test. I often hear how people can tell by looking that a certain image has that 'full frame look' or that smaller sensors look more 'clipped' or such. In some cases, I certainly agree. There are some instances where some cameras have a depth to the images that aren't present in other cameras. However, I don't think it's always so cut and dry...a lot depends on the lens, on the processing and on the photographer.
Look - this is just for fun...no agenda, no anything. I have no idea whether the results will be all over the place, or whether most people will get it correct. There is no 'gotcha' here, so if you are thinking "I'll pick them in reverse order, that'll make me right!" you'll probably be wrong. I am not showing crops because these cameras are from different generations, and noise isn't directly comparable between sensors of different generations. This is about the rendering....can you see a noticeable difference in rendering?
Each camera was mounted with a (very good) normal lens with f/1.4 maximum aperture. Shot to obtain roughly equal depth of field and at ISOs required to get the shot (so full frame was shot at f/2.8 and a higher ISO than Micro 4/3, which was shot at f/1.4 and a lower ISO...APS-C at f/2 and a middle ISO...yes I realized later I should have technically shot it at f/1.8, but there's a very minor difference there). Why? Because we've all heard "but if I want more depth of field we can always just raise ISO on the larger formats." However, for this test, try not to pick noise out, as it really has very little to do with what I'm after here...it's not a noise test. I tried to match framing as best I could.
The Micro 4/3 combo is the OM-D E-M5 with Panasonic Leica 25mm f/1.4. The APS-C combo is the Fujifilm X-M1 with Fuji 35mm f/1.4. The Full Frame combo is the Canon 1Ds Mark II with Sigma 50mm f/1.4. Yes, I wish I had a 5D III for the full frame...but I don't. High ISO noise performance aside, though, the 'look' shouldn't be that different. Or maybe it is. Yes, I realize that an indoor scene under artificial light doesn't stress dynamic range and a landscape scene may completely change the outcome.
But remember, this is all for fun!
Vote which order you think the images appear by format.
i'm voted by dof which could be wrong, but 1 - least dof (FF?), 2 - most dof (µ4/3?), 3 - in between (aps-c). scene dynamic range is limited and all the shots show a fair bit of grain but 2 has the most, 3 has the least, and 1 is in between.
But I'm going off Field of View mostly. The Sigma's known to be more like a 46mm than an actual 50mm and the second image looks to have the widest view, and the Fuji 35's Field of view is a bit tighter than the other two (at ~54mm-e) and the last image is framed the tightest
It's most difficult for me to notice format differences on people pix. I guess M43-APSC-FF because of the skin tones, but that's totally depending on processing so... Total WAG.
Can't go by framing. The sigma is shorter than the rest, but adjustments in framing we're done on the fly. DOF should be roughly equal as well due to aperture choice.
when you go out of your way to make all the shots look the same, of course it's going to be hard to tell them apart, but I don't really shoot that way, so, sadly, this test proves nothing to me except that if you try hard you can make any format kind of look like others... You can make Tofu tast like seafood, but it still doesn't even come close, I still want my FF NEX.
Thanks for the effort though.
You could have shot the APS at F1.8 and the FF with a 120$ 50/1.8 at F2.5.
Lightroom? Then I'd try to find another color profile for #1, skintones are really a bit too yellow.
If I've switched the Oly (3) and the Fuji (1), then I'd have to say that I'm really disappointed in the 35/1.4, since #3 has such nasty LoCA just after the focus plane. But I've never actually seen it used for near-focus stuff before.
LightShow wrote: when you go out of your way to make all the shots look the same, of course it's going to be hard to tell them apart, but I don't really shoot that way, so, sadly, this test proves nothing to me except that if you try hard you can make any format kind of look like others...
and you think this "test" is meant to prove something, because... ?
LightShow wrote: when you go out of your way to make all the shots look the same, of course it's going to be hard to tell them apart, but I don't really shoot that way, so, sadly, this test proves nothing to me except that if you try hard you can make any format kind of look like others... You can make Tofu tast like seafood, but it still doesn't even come close, I still want my FF NEX.
Thanks for the effort though.
I didn't really go out of my way to make them look the same. I stopped them down to get similar depth of field and them set the white balance to the same K value in Lightroom. That's it. These are straight out of Lightroom at defaults except for white balance. No crazy curves adjustments or such. If you wanted to take this picture with these cams, they were set to the settings needed to get the shot and nothing more.
And, as mentioned...nothing to prove here....it's just a fun exercise.
Without looking at anyone's responses I voted for the second line, which I guess would be Canon, Panasonic, Fujifilm. I'm basing my guess on character of transition to blur and also a bit on skin tones. I know the colors would be hugely affected by processing, and I've not used any of those exact camera models, but hey, it's for fun!
I'm guessing Fuji, Oly, Canon, mostly judging by tonal range. The third image looks the smoothest to me (least posterization and graininess) and I'm guessing that the m4/3 image would look a bit sharper (slightly more DOF) than the Fuji image. I wanted to see the voting results before voting, but apparently that's not allowed, so my guess isn't included in the statistic.
Maybe the first image is the Canon though and the 3rd the Fuji... maybe I stick with that.
Jman13 wrote:
I didn't really go out of my way to make them look the same. I stopped them down to get similar depth of field and them set the white balance to the same K value in Lightroom.
fwiw, you should probably have set white balance to the same point in each photo in LR rather than setting it to the same K values to minimize color differences. different lenses definitely have different color casts and sensors have color filter differences...
.
I guess M43-APSC-FF because of the skin tones, but that's totally depending on processing so... Total WAG. 
when you go out of your way to make all the shots look the same, of course it's going to be hard to tell them apart, but I don't really shoot that way, so, sadly, this test proves nothing to me except that if you try hard you can make any format kind of look like others... You can make Tofu tast like seafood, but it still doesn't even come close, I still want my FF NEX. 