A lot has already been written in the internet extolling the virtues of Fujifilms’ XT-1, X-Pro, and XT-2/X-Pro 2 jpegs so I am not going to show you scientific evaluations, but I am going to give you some thoughts on the latest results I have been getting from the XT-2, and a few comparisons.
Firstly, let me say that I was never entirely convinced by the XT-1 and X-Pro 1 jpegs and did not place them equal to what could be quite easily obtained from minimal processing of the Raw files. Secondly, let me also say that I would always shoot Raw and JPEG unless I was shooting purely for putting something onto eBay or a similar circumstance or platform. This is simply because no matter how good the camera’s processor is, there are times when you need to pull a bit more from the raw than it is capable of, and there are times when you need to be in control of the processes.
However, there are also many times when you don’t need to do the work on the computer if you put in some work at the time of shooting, and this is especially true to the jpegs coming out of the XT-2 (and I imagine the X-Pro 2 since they share the same sensor).
You can’t really see it in an onscreen format, you need to compare prints, and you also need to compare them at significant sizes to see any substantial difference, but rest assured, it is there.
The question is, do you make big prints? Or do you just view the image on screen.
Another question is, do you think about what you want from the image when you shoot it? Because, when we used to shoot with film, we had to. We had to select the film stock for the shoot, or carry two bodies with two film stocks. Shooting jpeg is to me, exactly the same; the decisions you make at the time are almost as permanent as you would if you were shooting with film.
Raw will always allow you to change your mind later, without degrading the image. Convert your raw to Acros and it’s like shooting Acros. Shoot in jpeg and Velvia and then convert to black and white and you will not get the same tonal ranges that are available in the raw conversion. This may not be a bad thing – it instills discipline in the shooting.
Today, I shot in both raw and jpeg as normal, and I had intended on deciding later on which shots would be converted to black and white. In the end, I loved the Fujifilm jpeg rendition of Velvia that I haven’t converted any of them.
This shot, originally intended to become black and white, I actually like in colour now I am looking at the in-camera shot Velvia setting jpeg. I would still return to the raw file IF I decided to convert to black and white later because I want to maximise the options. But, as it stands, in colour, I am more than happy with the result here.
For interest, this is the raw converted to black and white version below:
I love the saturation of the rust colours and the beautiful clear definition between the tones. I also love the sharpness of the images from the XT-2 with the XF WR 16-55mm F2.8 lens, and to be honest, I can’t tell which is which – in-camera or raw processed.
Shooting wider, with landscapes, the images are also just as crisp from the camera:
So, did the in-camera jpeg let me down anywhere? Well yes, it did actually. In a heavy contrast situation with my dark colour dog heading into dark shadows on the bridge the jpeg wasn’t great. Here are the three images for comparison:
Whilst I got a good recovery from the in-camera jpeg, the raw processed image is still superior and doesn’t loose as much of the correct colour tones during pulling the shadows. Patches’ tongue looks far more natural for example.
So, whilst there are advantages in speed and accessibility of the jpegs, for tricky jobs, or when you might want to do more with the image at the processing stage (tricky light, produce both colour and monochrome images, for example) raw still offers more.
However, if you really have what you want in the jpeg I can see no argument for not using it in the vast majority of small print/online situations.
NB. I know Patches is not as sharp as he should be and this image would be rejected except that it shows the limitations of the in-camera jpeg in grab shot situations.