Fuji X-E1 and Nikon D7100 some ISO comparisons


I took a look at the relative ISO performance of my Nikon D7100 fitted with a 35mm f/1.8 DX lens and my Fuji X-E1 fitted with an 18-55mm zoom. Both APS-C sensors, and neither has an anti-aliasing / low pass filter.

The following were shot at ISO 200.


The above comparison is with the D7100 downsized to match the X-E1, both files being processed from raw in Photoshop ACR.  Very close with the D7100 having marginally more definition.


Above was similar except the X-E1 file was processed in Aperture. A marginal improvement for the Fuji.


The third comparison was to take account of the larger pixel count of the Nikon and I upsized the Fuji file. Again the Nikon is sharper.

Below are two high ISO comparisons from out of camera jpgs. At ISO 1600 and 6400.



Advantage to the Fuji here, though probably by not as much as I would have assumed.

Finally to take jpg. processing out of the equation, here are the ISO 1600 files processed from raw.


Again (slight) advantage to the Fuji, though the D7100 is very close.

So, what to make of this. I think what comes across very strongly is not the differences between the two cameras outputs but how similar they are (As is the lenses performance). I had assumed that the Fuji would be better at higher ISO's, but the Nikon isn't that far behind. The extra pixels of the Nikon give a slightly sharper rendition at ISO 200, but again its nothing that a bit of unsharp mask couldn't nullify. 

My main conclusion from this, is that its great how sensor technology is constantly improving and APS-C sensors can now create files that 35mm sensors would have struggled to achieve not that long ago. 

Two things should be bourne in mind however. Firstly the majority of the images are with the D7100 files downsized to match the Fuji, so that does give a marginal advantage to the Nikon and secondly of course the Fuji is a much smaller and lighter camera. 

It really is a case of 'You pays your money...... etc.' and whether you want a DSLR or a high-end mirrorless / E.V.I.L / CSC, these days, what you pick is probably down to other factors. Certainly both of these are excellent cameras and there are very few photographic situations that they couldn't cope with, in terms of quality of results certainly. As is usual with these tests I do, this is probably the only time I will ever shoot with either camera at ISO 6400 and the ISO 200 results are much more interesting to me. The Nikon, of course has a genuine ISO 100 which is excellent. I did a post including that a while back and comparing the camera with the D800E - HERE. Plus you can download some raw files and make up your own mind - HERE.

I do testing like this to clarify in my own mind whether there are any significant differences between the cameras I use. Obviously shooting with them and editing I get a pretty good idea of what they are capable of, but these tests give me a good idea of what I might expect. None of this takes account of how each file will sharpen and respond to post processing and there are certainly differences there, but that is down to individual preference so I try to keep that out of these tests. 

Finally, none of this is done to prove a point or demonstrate one brands superiority over another. I do this with all cameras I own as a matter of course (usually when the weather isn't great of corse) and at some point I'm sure I'll be doing comparisons with the other cameras I have. This is after all heavy-duty pixel-peeping, and while there is nothing wrong with that, it has to be seen in context. In terms of the reproduction of images from either camera in the print media which I supply via the libraries I work with, I doubt whether it would be possible to discern any real differences between them, even at A3. Plus I suspect an A2 print wouldn't be that different either. I write again and again about my belief that camera outputs in terms of image quality are really not that much different to each other. My way of making money involves my images being pixel-peeped and its obviously in my interest to do the same. But even allowing for this there is still more similarity than difference. Composition. and relevance are much more likely to earn me money from a picture than the amount of noise at a specific ISO or the accuracy of the reds. And thats the way its supposed to be after all.