The strange case of the Sony a6000 and Fuji X jpgs with contributions from the GX7 and A7r.

In my post yesterday comparing the Fuji and Sony wide-angle zooms I noticed something strange. The Fuji OOC camera jpgs. were darker but required a slower shutter speed. I even updated the post and did another test. Same thing.

Someone on Google+ had spotted the same thing and this is our discussion.

Ricardo Hernandez
Yesterday 20:33
Sound image - the issue is that fuji for some reason is behind in iso rated sensitivity by 1/2 to 1 stop vs other cameras at each particular iso (the difference varies per iso).

This means that for example the fuji xf35 f1.4 in terms of light with the fuji will be quicalent to the Sony f1.8/f2.0. It does not seem to be the lens but the camera doing this.

So while you shoot iso 6400 on the Sony its more like iso 3200-4200 on the fuji. This needs to be taken into account when comparing.

My Pentax k-5 was about one full stop more sensitive at iso 3200 and up vs fuji.

This means the high iso performance of the fuji while not bad its not as good as it first seems and its in line with other current apsc. You still get a benefit for not having the AA. 
Yesterday 22:03
What source do you have for that? It certainly seems to be true but I've never seen anything about it before.

Ricardo Hernandez
Yesterday 23:02
This has been commented for a while now. Places like dpreview pointed it out in their tests and there where user comments but basically my source is myself- I just tested it for myself on an xe-1 vs the k-5.

You still get shallower DOF of say the xf35 at f1.4 but with the camera in terms of exposure it operates like an f1.8-f2.0 vs some other cameras.

Could it be each fuji lens has a T factor of 1/2 a
Stop to a stop darker? That's also possible but seems unlikely. One way to check that would be to use the same lens on the xe1 and another camera that is not fuji branded with an adapter. 

An f/1.4 lens is always an f/1.4 lens, it's just the light transmission is different. I found some posts that say the ISO is wrongly stated but that doesn't matter. What matters is that the Fuji needs to let in more light i.e a slower shutter speed, to achieve the same result as the Sony. This happens wide open and I doubt whether the maximum aperture isn't f/4 like it's supposed to be. So this means either that the shutter speeds are inaccurate, which again seems unlikely, or what I suspect might be the case that the sensor array needs more light to achieve the same result as the Sony.

On the Sony DSLT's and using the LA-EA4 adapter there is around 1/2 stop more exposure needed to get the same result as cameras without a fixed mirror. The Fuji sensor array has more green pixels and I have no expertise here, but it may be the case that this different configuration needs more light. These extra green pixels might explain the green foliage smearing as well, but again I'm guessing.

Whatever the cause it seems that Fuji cameras may be up to a stop 'slower' than Bayer sensor cameras in certain situations. As you stated before the implications of this are that these cameras aren't as good at high ISO's as people imagine, since the files need to be lightened in post production to achieve the same results as other brands. If so that makes them less useful than they would seem.
Sound - yes an f1.4 lens is f1.4 but not or light I the T rating is slower than the F.

I don't think its that the fuji sensor needs more light. I think its that fuji calibrated their sensor to underexpose perhaps having to deal with their dr calculations.

There's no particular reason why xtrans would need more
Light as xtrans is just a different color filter array. 
+Ricardo Hernandez 'I don't think its that the fuji sensor needs more light. I think its that fuji calibrated their sensor to underexpose' That's the same thing.
If you have a look at the jpg. files in the folder I've made available for download in this post - you'll see that the Sony jpgs. also have higher shutter speeds to produce lighter images at the same aperture than the Fuji X-E2. They also do the same thing in comparison to the GX7 files though the difference is less marked.

Whatever the reason, this gives the a6000 a significant advantage in light gathering and means that higher shutter speeds / narrower apertures / lower ISO settings can be selected to achieve the same result as the other cameras. This has all sorts of implications for noise performance and image quality and gives the a6000 a real advantage.
No that is not the same thing. One means the sensor properties themselves require light gathering a certain way and another that fuji may be not amplifying the signal a certain way post sensor capture for some reason

They are not the same thing one has to do with innate properties the other with what fuji decided to do to make their iso scale and shift the dr window of they did. 

Whichever way you look at it you have to let in more light via a slower shutter speed on the Fuji to get a result similar to the Sony.

ISO settings in effect are only a guide anyway and they are often not that accurate. I've just done another post on this and for that I've tried four cameras with shutter speed and aperture set manually. Again the results show the Fuji jpgs. to be darker with a lower shutter speed. I agree that this could be the sensor requires more light or that this is something Fuji are in-camera processing.

But the result is still the same. Fuji X cameras require more light to achieve a satisfactory result. And for some reason the camera? / sensor? is either less efficient at getting enough light onto the sensor and to process the image 'correctly' or Fuji are 'software interfering' with this.

I've now tried three different lenses on two different cameras and the results are pretty much the same. I suspect that all we are going to come up with are theories here, since I doubt Fuji will shed any light on this. Plus since DxO have never tested a Fuji X camera, there's no help there either.

It makes for an interesting discussion however and certainly helps me make up my mind as to what cameras to keep or sell.

This is interesting and if true means that the Fuji ISO performance is not as good as it's made out to be. It seems that others have noticed this too. So this morning I compared 4 different cameras and decided the best way to show this was to set all cameras to manual, thus bypassing any ISO discrepancy. This is what I got at 1/320 sec. at f/5.6.

As you can see the Fuji X-T1 file is darker and the Sony a6000 slightly overexposed. I then let aperture priority take over on the Fuji, used my eyes to work out a manual exposure for the Sony, set the aperture to f/5.6 on both and the results were dramatic.

In terms of the shutter speeds this is about a 11/2 stop difference. And the Fuji shadows are much more blocked out than the Sony. Both jpg. settings were set to standard and all the parameters set to zero. Now the difference is much more pronounced with the a6000 than with either the Panasonic GX7 or Sony A7r. Now I would expect some minor differences, but nothing like this. And it does call into question what is happening with the Fuji jpgs. Is it really the case that the Fuji X sensor requires more light or is this Fuji processing the jpgs. in camera? I really have no idea as to what is going on here and it also seems that the Sony a6000 goes the other way. I have noticed that my cameras jpgs. are slightly overexposed for my taste and I'm now using the +/- dial set at -2/3 stop. I keep the Dynamic Range Optimiser turned on which works very well. I should also mention that I keep the Fuji version of this switched on as well.

So if anybody has any ideas as to what might be causing then then please feel free to comment in the Soundimageplus Blog Readers Group. Certainly makes the a6000 a lot more interesting than I initially thought however.

  • All original material on this blog is © Please Respect That

  • 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.
  • For commenting, discussion, posting your pictures, links and articles - join the Soundimageplus Blog Readers Group on Google+