Fuji X better than m4/3 for noise at higher ISO's - Myth or reality?



By accidentally applying the wrong preset to a m4/3 file I was editing I found that an image I had shot at ISO 1600 was actually less noisy than I was expecting. Hmmmm... I thought. That's interesting. Plus there has always been this feeling nagging away at me that while Fuji X raw files are less noisy at high ISO's, particularly with the OOC jpgs. they are softer, unless I use Photo Ninja or Iridient Developer. But then using either of those does make the files they process look noisier both in terms of luminance noise and colour noise.

Now I'm VERY interested in the Panasonic GH4. Every example of test shots and video footage I see makes me think that this could be a special camera, particularly since my plan is to shoot a lot more video in the future. However, almost every comparison and opinion (including mine) is that the Fuji X is much better at handling high ISO noise than m4/3 cameras. And certainly comparing like for like, using the same settings, that is what I have found. But then I got to thinking, what about if I don't compare like for like? What about if I used the settings I have come up with to get very clean, sharp m4/3 files at low ISO's and adjust them to see if I could come up with some settings that 'equalised' the results between my Fuji X and m4/3 cameras with regard to the noise / sharpness equation. So I did some test shots.


Above are two ISO 1600 jpgs. from my Fuji X100s and Panasonic GM1 + 17mm f/1.8 lens. Clearly the GM1 file is noisier, but also sharper.


In the example above I sharpened the Fuji file which worked, but it is noisier.


Next, I reversed the procedure. The Fuji file is as before the OOC jpg. but this time I applied noise reduction in Photoshop to the GM1 file. As you can see, it's softer certainly but in terms of noise and the overall look not that dissimilar to the X100s.

Then I had a look at some raw files.


Above is a comparison of identically processed raw files. The GM1 file is clearly a lot noisier, but again it's sharper.


My final comparison was with the Fuji raw file left as it came out of Adobe Camera Raw, but with noise reduction applied to the GM1 file. This brings the two cameras much closer together. I still see a cleaner result from the Fuji X camera, but with very careful processing and the application of Photoshop noise reduction this is better than I would have thought. The GM1 file does loose a degree of sharpness, but in reproduction terms for the markets that I sell to, this is negligible. 

Encouraged by this I had a look at some files I shot in museum last summer. I haven't done anything with most of them since I had some refused by a picture library as being too noisy. Even using a Nikon f/1.8 prime fitted to my GH3 with a Metabones speed booster I was still shooting wide-open at ISO 1600. The raw files and the OOC jpgs. for that matter are pretty noisy, but applying what I had discovered above, I got a much cleaner, though softer, result.


There is no doubt that the Fuji X-Trans is better for this kind of work. It's obviously easier and quicker and works 'straight out of the box'. But it wouldn't be difficult to set up a one click preset for getting decent high ISO results from my m4/3 cameras. I just have to get the settings right. 

There is however the fact that Fuji are using in-camera software and post-processing 'background' presets to keep the high ISO noise down. And to my mind this is more important that the fancy sensor array they have come up with. m4/3 and Panasonic in particular have always prioritised image sharpness over low noise and it's not that difficult to come up with some settings that produce similar results to what happens with my Fuji files. To what degree this can be achieved however needs more testing.

What I'm now interested in doing is seeing if I can get results that I'm happy with, using my m4/3 cameras, at what are becoming my preferred settings for my Fuji cameras. I.E. ISO 400 > 800. These are settings I've always avoided with my m4/3 cameras, but I want to see if it's possible to work this way with those. I'll report back when I've done that.

As ever I have to issue my usual disclaimer that this is under no circumstances intended to be some kind of definitive, objective analysis of how these different cameras perform. This is limited to what I shoot, how I shoot and what I shoot with and is something that I'm experimenting with to see if I can get better results from my gear. What I want from a file is not necessarily what others want. So please bear that in mind. 


N.B. to see more on the cameras and lenses featured in this post click on the relevant labels (tags and keywords) at the bottom of this post.
 
All original material on this blog is © Soundimageplus.  Please Respect That  

LINKS AND SOCIAL MEDIA    For commenting, discussion, posting your pictures, links and articles - join the Soundimageplus Blog Readers Group on Google+ 
about soundimageplus 
follow by email 
follow on twitter
follow on facebook