I’ve been thinking about how to go about reviewing and writing up the Fujifilm X-T2 and it is clear to me just how much there is to say. For now, I am going to write my initial impressions, covering a wide range of topics fairly superficially (OK, perhaps 7,000 words isn’t that superficial). I will then delve into some of the more important sub-topics in more detail in subsequent pieces. If I make mistakes or there are oversights, please don’t shoot me. I am new to the Fujifilm system and it is inevitable that I will miss out on some features and perspectives that experienced users may consider important. The upside of this is that I am coming to the camera (and system) with fresh eyes.
As many of you know, I have spent years shooting with Leica M cameras, so I value portability, tactile qualities and optical excellence. I’m also a huge fan of the Sony FE system and I have shot Canon for years (all the telephoto portraits in the series Russians and Royals and Afghan Heroin: Not For Export were shot on Canon, for example). I also love the APS-C Ricoh GR and appreciate M43 cameras, so I am not coming at this with any kind of format prejudice. I will try to cover as many angles as I possibly can and, if you feel something is missing as this series of articles unfolds, let me know and I will try to fit it in.
I won’t go into minute detail when it comes to he bewildering array of features (or their set-up), because other sites are better positioned to do that. What I will aim to do in this piece is provide a qualitative overview of the camera itself, but also to think a little about how this camera sits alongside potential competitors.
Before I do so, I’d like to say a huge thank you to the team at London Camera Exchange, Chester. They’ve been enormously helpful (not to mention patient) and I can thoroughly recommend them to any Brits looking for new equipment. I have purchased all sort of items over the last few years from the LCE website (which can be fulfilled from a variety of branches) and from the Chester store itself. I don’t make a habit of referencing stores on this blog, but it would be inappropriate for me not to do so in this case. Credit should be given where it is due!
Fujifilm X-T2 Build Quality
This is on broadly par with the Sony A7 II generation cameras, with the Fujifilm camera having an edge in my opinion. The metal alloy bodies have the same general feeling of solidity, but the buttons on the Fujifilm XT-2 are decidedly better. The controls are more positive and the diminutive control dial on the back of the Sony is a weak point. Instead, the Fuji has a pad that feels more ‘decisive’ and more solid. The area where I feel the Fujifilm has a big advantage is in the shutter button. It is metal and its operation far superior in the opinion of my own trigger finger (more on this later). Overall though, there isn’t a huge gap between them in terms of which one would better survive a fall from the back of an elephant while tiger spotting in India. Neither the Fujifilm XT-2 nor the A7 II feel as robust as a Canon 5D III or IV, but there are very few cameras that do. Besides, the Canon is nearly 3 x the price (UK) of either Sony or Fujifilm.
The grip on theFujifilm XT-2 is less deep than the Sony A7 II and preference will be very personal. Both require a slightly different grasp of the camera, but I find each to be comfortable and secure. Most of the Fuji’s controls fall nicely to finger/thumb and here I found the X-T2 a little easier to operate the rear buttons with my eye up to the viewfinder. The Sony grip is better suited to larger lenses (of which there are now many in the Sony FE line up), but it is worth noting that Fujinon X lenses are appreciably smaller and lighter on average.
On top of this, there is a new Power Booster for the Fujifilm X-T2 (that offers some notable performance improvements), which I have here with me. Not only does it significantly boost the speed of continuous shooting (up to 11fps with mechanical and 14 fps with electronic shutters), it carries an additional two batteries, swells the grip a touch and alters the balance with longer lenses. More on this later.
The Fujifilm XT-2 seems to have hit a sweet spot in terms of fusing digital era controls with mechanical dials, personally speaking. In short, I love the layout and feel it is extremely practical. All the usual controls are there, with plenty of function buttons and options for button customisation. The joystick is brilliant and allows very quick and precise navigation of focus spots and they can readily be adjusted for size. The ‘cursor’ seems to skip around more quickly than on my 5D III and the whole process feels responsive and accurate. The only shortcoming I have found here is that the joystick could do with being higher up. Where it is now means a slight shift in grip to find it with my thumb and move the focus point. Not a huge problem, but it isn’t quite where my thumb would prefer it to be. While dials are not illuminated, as a top plate LCD is, during the majority of conditions it does make it very easy to see where you are at in terms of settings. It is a matter of taste, but I prefer this layout to the A7II and Canon. It may not be rational, but it really does feel ‘right’, for myself at least.
Weight is important and at approx 500g with card and battery, the Fujifilm XT-2 is similar in weight to the Sony A7II & A7R II. With Fujinon primes attached, including faster examples like the 23mm f1.4 R, I found it a very comfortable camera to hold, shoot and walk around with. It feels agile, which I suspect is a combination of the fairly light weight of the lens (ensuring balance is not too far forwards) and the more ‘sporting’ grip on the body (as compared to the A7 II, for example). I will look at the balance with longer and heavier lenses when I review the individual lenses, but with the sort of optics I consider to be most important (small to medium primes), I feel the Fujifilm XT-2’s handling is superb.
The question some of you may ask is, ‘why on earth buy an APS-C camera at a given size and weight, when I could get a Full Frame like the A7 II. It also costs about the same too?’. It is a fair question and I will answer it in more detail as I go along, but the short answer is this: performance benefits.
Fujifilm X-T2 User Interface & General Operation
I find that my ability to operate the camera, find things in the menus, set the camera up and go about the business of photography is nice and easy. It is a far cry from the original X100 that regularly left me bewildered about how to change a setting. Some things were neither easy to find nor easy to remember, even after you’d looked it up (for the fifth…. or tenth time). The current menu (and quick menu) system is excellent and I can’t say there is anything I have wanted to change that I can’t change or find and find again. It is far superior to the current Sony system (which is in the process of being overhauled and improved). Fujifilm has a well-deserved reputation for listening to their customers and it shows.
Except there is one niggle. I love the C1, C2, C3 dial system on the high end Canons. Yes, you can set up custom settings on the Fujifilm XT-2 just fine, but the Canon system is IMHO the most useful system of all. It allows you to change a whole slew of performance parameters with one click of a physical dial and then back again with another click. I will admit that I need to spend much more time learning about the X-T2 and its set up, but I can’t yet see how one can change such a broad range of settings with such a small amount of user input. This sort of thing matters a lot to some people (wedding photographers, photojournalists), but not at all to others. What IS excellent about the Fujifilm XT-2 is that whatever you do wish to change is very straight forward and transparent. You aren’t going to find the camera behaving in an unexpected way, yet unable to figure out why.
OK, another small niggle: rear focus. Yes, you can set the Fujifilm XT-2 to rear focus (I’ve set AF to the AE-L button) and removed AF from the shutter button. However, the button itself is almost flush with the body itself. This means that it is harder to locate and positively press than with a Canon or Sony. Interestingly, Fujifilm got this same button absolutely perfect with the X100F. To look at the cameras you would not see much of a difference, but in use, it is better implemented on the X100F. With the X-T2 it is possible to press it, but not quite hard enough, or with enough ‘finger flesh(?!)’ to set AF in motion, because it does not sit quite proudly enough. I would imagine that Fujifilm was worried about people accidentally pressing this button, as it is where the thumb naturally falls, but at the end of the day it does work. It just isn’t quite as refined as it could be. If you do not use rear focus, but do a lot of street photography, documentary, reportage (or weddings?), you may wish to give it a try. In my opinion it can be a far superior way of operating the camera for reasons that are perhaps best explained in a separate article (let me know if you want me to write about this).
The dial locks are simple, very positive in operation and work perfectly.
Tilt & Swing screen: Thanks Fujifilm, this works great. most photographers want one and this one is nicely implemented. The Canon 5D IV does not have one and IMHO that was not a wise decision.
How Does the Fujifilm X-T2 Feel? This is something that is personal, but as many others have said before me, I find that the Fujifilm X-T2 feels more like a ‘proper camera’ than the Sony A7 series. This is a point often made by users and I agree. The Sony is OK, to me, but they are not ‘fun’ to operate and they do not feel much like mechanical cameras of old (they are however, superb imaging tools). The Fujifilm X-T2 is not like a Leica M either, but it is noticeably ‘nicer’ to use than the Sony A7II bodies (to me). All of this is true before you press the shutter button and even clearer afterwards. The shutter release is vastly superior to Sony’s on the A7 II/ A7R II and in my opinion quite superior to the Canon system: it is crisp, decisive and perfectly weighted. You know precisely where you are with a half-press and you can gauge the point of release accurately. Come to think of it, I think it is among the nicest shutter release I have ever used on a digital camera.
X-T2 Shutter: It is quiet by Full-Frame standards and much quieter than the Canon 5D III in normal operation. It is also noticeably quieter than the Sony A7 II, in part because the X-T2’s sound is softer. The A7 II sound harder and lasts longer. This isn’t a surprise considering the Sony has a larger shutter to cover the larger sensor. The sound recording below should give a good idea (Fujifilm X-T2 first).
The Fujifilm XT-2 has a totally silent electronic shutter mode that neither the A7 II (nor 5D III) has. In general the shutter/release feels very responsive, with no evident shutter lag. In this regard it feels much better than the Sony A7 II. Overall, the sense is of a camera that is ‘ready to go’ rather than one that is a bit sleepy (A7 II).
General Operating Speed? The Fujifilm X-T2 is a quick camera to operate. It feels snappy and responsive in every way, whether you are moving through menus, operating the joystick or focusing the camera. It is not a small leap up from the Sony A7 II, but a large one. I think it was Ming Thein who said that even the A7R II felt like everything was moving through molasses. While this may sound unkind, it does describe the general lethargy of the A7 II generation compared to some of the current competition and certainly high end DSLRs. Now the X-T2 has very much joined high end DSLRs in terms of responsiveness and it means the performance envelope of the camera has moved further into DSLR territory. If capturing split-second moments if your profession, it is not hard to see why the Sony FE cameras are (thus far) relatively poorly represented. Weddings are arguably one of the most arduous environments for a camera in terms of flexibility and general speed of operation. It is perhaps telling that these areas are still dominated by Canon, Nikon and that Fuji has made inroads with the X-T1, which have found favour with quite a few wedding photographers. There is much more to it (ISO, image quality etc), but the Sony camera just feel a bit wooly and vague. Take those Sony cameras out for general travel and landscapes and it is a very different story. Mount Snowdon tends to be fairly indifferent to speed of AF or crispness of shutter release. In terms of outright imaging performance, the Sony cameras are tough to beat.
Fujifilm XT-2 Viewfinder
It’s excellent. At 0.77 magnification, it’s just noticeably larger than the Sony A7 II (0.71) finder and effectively the same as the A7R II (0.78). Compared to the Sony A7 II, the X-T2 finder seems quite a bit brighter and a ‘clearer window into the world’ than the A7 II. While resolution is similar (around 2.3m pixels), the X-T2 has better eye relief and (for me) a better eye position. It all feels a bit more ‘open’. You’ll have to compare for yourself, but to me the Fujifilm XT-2 has a clear edge here. The XT-2 finder has more contrast by default, but this is easily adjusted by adjusting your JPEG default settings. I prefer mine to be lower in contrast and more like real life. The default contrast settings for the Sony FE cameras’ EVF is a bit better for me in this regard, but the Fujifilm EVF seems better in every other way and can be adjusted in any case. In terms of information provision and ‘cleanness’ the X-T2’s EVF is a pleasure to use.
Fujifilm XT-2 Autofocus
I have not conducted detailed testing here, although most readers will know that the Fujifilm XT-2 has a reputation for excellent tracking. What I can confirm is that here is significant variation in AF performance depending on which lens you use. In general, AF speed seems very good. In fact it is extremely good with faster focusing lenses (which include the 18-55mm kit lens). While it does not rival the fastest M43 cameras, it seems to me about as quick as the Canon 5D III, or possibly a bit quicker. With older lenses, such as the 23mm f1.4 R, it is still fairly quick, but there is noise that users of Canon USM, Sony FE and similar optics may not be expecting. I’d not describe it as loud, but there is a clear noise as the lens components ‘shuffle’ into position. This movement sometimes feels a bit ‘vague’ and does not convey an impression of accuracy because you feel something moving rather than going from A to B in one sensation. However, accuracy is very good from what I have seen. Even the slowest lens I have used to date (Fujinon 56mm f1.2 APD) has been fairly quick and assertive in everything from good to moderately poor light. I will be testing this further, but in similar conditions, lenses like the Canon 50mm f1.2 L are without doubt less accurate (and no quicker) and the 85mm 1.2 L II is significantly slower. Both of these fast Canon lenses focus with best consistency on my Sony A7 II body and as mirrorless users will be used to, focus accuracy is one of the strengths of focusing directly off the sensor.
Overall, single shot AF is quick and accurate with the Fujifilm XT-2. I would say that older lenses would really benefit from an update to bring them up to the standard the new bodies are setting. I’m sure that when I get around to testing some of Fujifilm’s fastest lenses, I’ll be even more impressed than I was with the 18-55 f2.8-4. The kit lens is effectively silent. I understand that the 16-55 f2.8 and 35mm f2 WR (and similarly up to date lenses) are both silent and quicker still. I expect to find out soon.
Preliminary testing of AF tracking has been very encouraging. It is vastly better than the regular Sony A7 II and seems at least as good as the Canon 5D III (and at a higher frame rate). I will check this out in more detail and will focus more on the kind of tracking that might be useful to documentary/reportage photographers, rather than the sporting or birding fraternity. For now, I already feel confident that it would be more than good enough for me.
Fujinon X Lenses
I will discuss specific lenses in more detail in future articles, but here are some thoughts thus far:
The 18-55mm f2.8-4 kit lens is superb by ‘kit lens’ standards. On the copy I have used, corners are not perfect wide open (but most of the frame very nearly is) and there is a touch of decentering visible at the wide end, but it clears up quickly upon stopping down. By f5 the lens is in its sweet spot for the most part. Like many zooms, at longer focus distances it can suffer at the edges (probably due to field curvature); however, I need to spend more time with this lens to comment in any more detail. Let’s remember it is a very small optic sporting a fairly fast aperture range. I feel it makes the most of the compact form factor that X cameras offer, while offering very good performance. It is a lens I suggest taking care not to overlook, if you are buying into the X system. With stabilisation thrown in, I know I would use it a lot, especially for travel and general walkaround shots. Construction is not Canon L, but it is very good for a kit lens, with lots of metal. At 23mm, as good as it is, it does not compete with the below lens however.
The 23mm f1.4 R s a lens that will be important to many people, because it is the fast 35mm equivalent for the system. Initial usage suggests wide open performance is excellent and stopped down to f4 or so it is effectively ‘perfect’. f2.8 is very close. It will be interesting to see how it compares to the Canon 35mm f2 IS, because this is arguably the direct competitor in the Canon system. Sony does not have a direct competitor, there being no f2 lens in the Sony line up. You have to choose between the very small and slow 35mm f2.8 Sonnar and the fast, expensive and enormous 35mm f1.4 Sony Zeiss Distagon. AF is ‘good’ on this lens, but it isn’t as quiet, quick or decisive as the kit lens. If Fujifilm were to update it to WR status with silent focusing, it would be a very nice upgrade, but the optics themselves are already fantastic. Not all 35mm equivalent lenses are nearly as good as this at the edges and corners. As I say, at f4, the corners look all but indistinguishable from the centre. Despite its speed, it is nice and light, yet solid (metal). It feels like the X-T2 and 23mm f1.4 R were made for each other, which I guess they were….
I am also testing the 56mm f1.2 APD rather than the regular R version. I am doing this because the APD version, from the examples I have seen, offers the most consistently excellent bokeh of the two. Sometimes the regular R looks almost as good and on rare occasions a hair better, but the APD never seems to produce anything other than superb bokeh. At times, the regular R can produce ‘energetic’ bokeh while under the same conditions the APD seems to pass with flying colours. For this reason, I would consider it to be the best benchmark against which to compare other lenses from other systems. It will be fun to see how it fares next to my Canon 85mm 1.2 L II and especially the Zeiss Batis 85mm f1.8. A 56mm lens for APS-C produces the same subject magnification as an 85mm on FF, which means you have the same sort of working distance for portraits. However, a 56mm f1.2 on APS-C is roughly the same as an 85mm f1.8/f2 in terms of visible background blur.
From initial usage of the 56mm APD, I can already say that it is amazingly sharp at f1.2, focuses better than expected, it is very accurate and backgrounds are indeed very good. As I say, there will be much more from me on XF lenses. <<— Since writing the last sentence, I have done an initial side by side test with various backgrounds and distances and feel happy sharing my preliminary conclusion. Compared to the Zeiss 85mm f1.8 Batis (which I own and find to be an astonishingly good lens), the Fujifilm 56mm f1.2 APD is every bit as sharp wide open, produces a different look (a bit gentler?) and (to my taste) has consistently superior bokeh. It’s not a massive edge, but it’s nice to know that this very expensive lens does seem to produce slightly more pleasant portraits than the very well-regarded Batis. Colour rendition out of camera is vastly superior.
Initial lens handling thoughts…. The 56mm APD balances very nicely and feels noticeably more compact (and a bit lighter) than the comparable Sony A7 II and 85mm Batis combination. The Fujinon X lens range is also consistent. All (except the 27mm pancake) have aperture rings close to the body (someone correct me if I am wrong) and a user of any given lens will find their way around a new lens in no time. The same cannot be said for Sony FE, where we have some lenses with knurled focusing rings, some with smooth ones, some with aperture rings, some without etc. This is not a show stopper of course, but the Fujinon range’s consistency is definitely a big plus for people shooting under time pressure. So far, I very much like what I see. They’re predictable in use, have similar ‘looks’ and they’re optically very good.
What About the Larger Fujinon f2.8 Zooms?
This is an area where some feel the Fujifilm system ceases to provide the sort of advantage that the smaller primes do, or rather that the ‘benefits of the small Fujifilm form factor is rendered pointless’. I’ve heard quite a few people say this, but I suspect it has more to do with just how sweetly the bodies handle with the smaller primes than how ‘huge and heavy’ they are with the f2.8 pro zooms Personally, I feel that this is probably just a matter of preference and what individual users are comfortable with (and why they are drawn to the system in the first place). Here are my thoughts about the ‘crop factor’, lens speed, size and utility. In order to really think this through, the size of the sensor is really important.
Generally speaking, when you convert the focal length for an APS-C lens to FF (such as a 23mm providing the same sort of framing as a 35mm on FF), you need to add about a stop to the speed of the lens to give an equivalent depth of field. This means that a 23mm f1.4 shot wide open on APS-C will provide a similar depth of field appearance as a 35mm f2 shot wide open on FF. The same 23mm at f4 will be similar to the 35mm FF lens at f5.6. From what I have seen in the many comparisons I have seen online, the factor is a touch longer i.e. a 56mm at f1.2 on APS-C will look similar to an 85mm FF lens shot at f2, rather than f1.8, however, the details of this I will look into in more detail in practical tests. In terms of light gathering, the f1.4 APS-C lens still lets in more light than the f2 FF lens (t lets in a stop more, of course). This means higher shutter speeds for a given ISO, but with comparable apparent depth of field. This has pros and cons and I will discuss this more later.
Enter the f2.8 zooms. Fujifilm’s 16-55mm f2.8 WR lens is the equivalent of roughly 24-85mm f4 on FF in terms of subject magnification from the same shooting position and in terms of apparent depth of field. While a small increase over the 70mm top end of many standard zooms, it does make a difference. After all, this is why some vastly prefer the 24-105 L over the optically superior 24-70 L offerings.
The problem, according to critics, is that the Fujinon 16-55 f2.8 WR is quite large and heavy relative to the X-T2 and other Fujifilm bodies. While this is true to an extent, at 600g, the Fujinon is still a fair bit lighter than a Canon 24-70 f2.8 L II (800g) and very much lighter than the admittedly stabilised Nikon 24-70 f2.8 E VR, which weighs a whopping 1050g. The issue is perhaps more due to the much lighter Fujifilm bodies they’re attached to and what this means in terms of net balance. Above are some size comparisons, obtained from www.camerasize.org.
Body plus lens weights are a different picture altogether.
Fujifilm XT-2 + 16-55 f2.8 = 1100g
Canon 6D + 24-70 f2.8 L II = 1600g
Nikon D750 + 24-70 f2.8 G VR = 1800g
However, whereas the DSLRs have a big grip, the X-T2 does not. What if we put a 375g Power Booster on a Fujifilm X-T2?
Now it weighs 1475g, which is a touch lighter than a Canon 6D combo, but we have 11fps available in mechanical shutter mode (14fps with electronic). This is double what the 6d can muster and I have a feeling tracking will be noticeably superior with the X-T2 as well. Yes, a 6D II will be coming, but it isn’t here yet.
Fujifilm XT-2 + 50-140 f2.8 = 1500g
As above, with power booster = 1875g
Canon 6D + 70-200 f2.8 L II = 2300g
So, as you can see, even with the power booster on board, there are weight savings, albeit not spectacular ones. The booster does, however, help to re-balance the camera by adding heft behind the lens.
There is of course an argument that the f4 range of zooms is the more accurate comparison and in some respects this is true. In severe light and wide open, a Canon FF set up with the 24-70 f4 L will give you comparable depth of field to the Fujinon 16-55 f2.8 and the larger sensor may allow a higher ISO to be used to provide the same sort of working shutter speed as the Fujifilm set up. However, when light is good and ISO is not a factor, the Fujifilm set up will allow for faster shutter speeds at wide open and the same ISO. There are lots of twists and turns in these arguments and it all depends on perspective. I don’t think there is a winner, because it really depends on what is important to you.
Overall, I would say that the benefits of the Fujifilm system do seem to shine brightest with the small primes attached. However, not everything has to be binary. You can still enjoy the benefits of the small and light primes on these compact bodies, while attaching a fast zoom when that suits your needs at a given time. We tend to buy into a system and I feel that in overall terms, Fuji has done really well here. While Canon has a range of excellent f4 pro zooms, Nikon arguably lacks the wide zoom (the 24-120 being a somewhat weaker lens, due to the longer range). Sony has a range of f4 zooms (Sony Zeiss 16-35 f4, 24-70 f4 and 70-200 f4G), but there is a fly in the ointment here: the crucial 24-70 f4, which for some users is too weak in the corners at 24mm and above 60mm. With f2.8 zooms attached, the Sony system of course brings lenses every bit as heavy as Canikon (and in some cases heavier) to bodies that are as light as the XT-2. Arguably this will result in the worst balance of all.
In general terms, one thing the Fujifilm lenses (I’ve used to date) manage to pull off is excellent resolution while still appearing smooth, or at least lacking in harshness. The primes also seem perfectly proportioned for the bodies too. This isn’t surprising, because there is an holistic philosophy behind the Fujifilm X system. With Sony FE, despite the many wonderful lenses, it is a bit all over the place due to the focal length and aperture ranges being raggedly carved up between Sony, Sony GM, Sony Zeiss, and ‘true’ Zeiss lines. The Fujifilm range may not be as extensive, but it is homogenous and covers pretty well everything a user is likely to need. I am sure there are people who would like long telephoto primes for wildlife, but I venture this is a tiny minority of users at the present time. With the Fujifilm XT-2 showing such excellent AF performance and tracking, that may of course change.
I am very impressed and look forward to seeing what some of Fujifilm’s most recent lens offerings have to offer.
Fujifilm X-T2 Image Quality
The X-Trans Factor: I’ll come right out and say that for some time I have viewed the X-Trans filter array with skepticism. There are two main reasons for this:
- I believe the issue of moire is not really a serious concern these days. As sensor pixel densities have increased, moiré has become less and less of an issue. Many camera models are now offered without Anti-Aliasing/Low Pass filters at all and these models often sell in greater volume despite costing more than the models with low pass filters enabled. If moiré was a serious concern, this would not be the case.
- The X-Trans filter has come at a cost. I suspect every reader is aware of the fact that there are variations in how well different RAW converters can de-mosaic the files. Some don’t even support X-Trans files at all (DxO, being just one). For me, and I suspect many others, continuity of workflow is very important. When the camera marketplace has so many great cameras for sale, why bother with a camera that might require you to change to a different processing platform? Not only is this disruptive, but it also has a financial cost.
It will take me time to work through these issues, but I will touch upon them further along in this post.
It’s worth noting that there have been some developments over the last few years: Adobe has a better handle on X-Trans files than ever before (as many users have confirmed) and there are fewer reported issues with odd textures appearing in fine branches and detailed green subjects. With the 24MP X-Trans III in the Fujifilm XT-2, I was very interested to see how it performed. Naturally, the first stop was the processor I use for every single other camera I own and have eve owned and used: Adobe Lightroom. Whether you love or loath Adobe LR, there is no denying that it is used by far more people than any other. So, some very early thoughts are as follows:
Resolution is excellent overall. RAW Sharpening needs to be approached very differently to cameras using Bayer sensors and the general advice seems to be: use sharpening of 25-40 max, radius of 0.8 to 1.5 (some use more) and detail of 50-100. This seems absurd, but it…. well, it works. The end result is a natural looking level of sharpening and loads of fine detail.
I did notice that some slightly odd things happened with fine twigs in trees; however, I saw this in out of camera JPEGs for which I did not have RAW files at the time. It also occurred in asphalt on a road, where slightly odd looking patterns appeared. I have since seen that noise reduction and sharpening (applied in camera) to the X-T2’s JPEGs can compromise the finer details, when those files are inspected in great detail. This is true for all manufacturers though and JPEGs rarely if ever match well-processed RAWs in this regard. I will look into this in more detail and report back; however, compared to some files I saw from 16MP X-Trans files some years ago (also processed in LR), the issues now appear to be very much diminished. It seems the latest Adobe RAW conversion algorithms and the 24MP X-Trans III are getting along pretty well. For now, what I can say is that the detail is certainly there and I have not seen any serious problems to date, including on grass and fine leaves and twigs. RAW files have lots more bite from the outset than, say, a 5D III or A7 II file. This is presumably due to the lack of a low pass filter.
As part of my future testing, if I generate any problem RAW files, I will look to see if other convertors do any better. Ideally, I will try to compare to my Sony A7 II. While Full Frame, at least it shares the same megapixel count.
Dynamic Range (DR): The base sensor is assumed to be of Sony architecture. Need I say more? Just in case I do: In short, I see bags of dynamic range (DR) and superb malleability. While a 645Z or D810 at base ISO will be better, the X-T2 files are about as good as it gets in the APS-C world and upon casual inspection, no worse than the A7 II. I have not tested a Canon 5D IV, but the files from the X-T2 are in a different (and better) galaxy to those from the 5D III in terms of DR, flexibility in post and banding. It isn’t even close, despite the difference in sensor size.
Noise: This is always going to be an interesting one, because Fujifilm is known to use a different ISO scale. This means that for a given scene, aperture and shutter speed, the Fujifilm will assign a different ISO for a given exposure than a Canon or Sony. The difference is about 1/2 to 2/3 of a stop, so a Canon may show it is using an ISO of 1250 at f8 and 1/125th, for example, while the Fuji will show it is using ISO 2000 (or thereabouts). I have not tested ISO performance in detail, but I have shot some basic test scenes and roughly taking into account the difference in ISO scales, my initial impressions are that the X-T2 is nudging some Full-Frame cameras in high ISO performance. The JPEGs are a huge improvement on early Fujifilm JPEGs and do not show waxiness or weird textures. Naturally, at 12,800, the JPEG engine is working very hard to deal with noise and detail, but I’m struggling to see any real benefit to the larger (albeit older tech) sensor in the 5D III. The Fujifilm X-T2’s high ISO RAWs clean up beautifully in LR and can be made even better with careful noise reduction. I see excellent (noticeably superior) colour retention compared to my 5D III, with the X-T2 looking better at a listed 12,800 than the Canon at 8,000. The Fujifilm also has an ace up its sleeve – see next paragraph.
Film Simulations: I never thought I’d say this, but these JPEGs are plain excellent. Just don’t go near ‘grain simulation’ which in my opinion is unlikely to find too many fans. Without grain simulation enabled, the camera just applies the colour profile, sharpening and noise reduction. With grain effect enabled, it tries to add grain to make it look like a corresponding speed of film were used, so you can imagine what ISO 12,800 looks like! Classic Chrome simulation is a favourite and Velvia will have its fans (including me, at times).
Then there is Acros, which is very nice at low ISOs and tremendous above. At a listed ISO of 6,400 and 12,800 the noise produced is wonderful (by digital standards) and is vastly superior to a 5D III file at roughly half a stop slower ISO. While the Canon produces noise that increasingly resembles static, with white luminance noise in deep shadows that is a pig to remove (a bit like the Leica M9 Monochrom at 10,000), the Fujifilm XT-2 X-T2 produces noise that looks much more like film grain on an ISO 400 film. Acros (with or without a filter applied) is going to be an absolute boon to anyone looking to produce a quick ‘n dirty high ISO file that does not need masses of work to remove that ‘noisy’ feel. Assuming the shot suits the reduced DR of a JPEG (compared to a RAW), the out of camera high ISO Acros JPEGs look more like film results than anything I have ever seen come out of a camera. For some reasons they are very much better than the camera’s already very decent high ISO colour JPEGs. It seems that Fujifilm has done something very different with the grain/noise for these Acros JEGs, resulting in a quite different feel to their colour brethren. I also wonder if the X-Trans filter is also helping to make noise/grain appear more random. Whatever it is, it works. Yes, there are many great film simulations out there, but this camera punches out fantastic B&W JPEGs as standard, especially at high ISO.
Fujifilm XT-2 Colour strikes me as a really good balance between natural and accurate. In direct comparison to my Sony bodies, I find myself preferring the Fujifilm pretty well every time. If you’re prepared to work on all your RAW files that not much of an issue, but if you have hundreds or thousands to work through, it is a very big deal. I am already very impressed by skin tones, which appear absolutely smack on, but also with the greens. Whereas Sony tends to produce quite a rich and dark green, Canons tend to produce quite a bright glowing green, which I sometimes find a bit unnatural. I shoot mostly B&W, as readers will know, but from what I am seeing, Fujifilm seems to have just the right balance for natural yet appealing greens.
Fujifilm XT-2 First Impression?
Extremely good. This is a camera that brings so much together in one camera. It is a much quicker/snappier camera than a Sony A7 II or A7R II, yet has a larger sensor than M43. It offers the sort of tracking many advanced DSLR users are used to, yet is much smaller and lighter. It has some of the tactile appeal of a ‘real camera’, yet offers a host of modern features. If you don’t need to upgrade a camera, there is often good reason to just sit tight. However, I feel that, as a generalist system camera, the X-T2 really does remove a good number of the reasons why a person thinking of upgrading might feel compelled to stick with their DSLR system. I can’t think of a single area where my 5D III meaningfully outperforms it. On the flip side, I can think of a whole load of areas where this camera very much outperforms the 5D III (and even the 5D IV) at a *much lower price point*. On top of this, and perhaps even more importantly, it is as enjoyable to use as it is good to use.
I will be back with more detail on image quality (including more detailed peeping regarding the X-Trans array), lenses and of course comparisons with other cameras. I will also discuss in detail what I think the Fujifilm system looks like and how it may stack up against Canikon, Sony and possibly even Leica for a variety of applications. I will of course come at this primarily from a documentary, reportage, travel and landscape point of view, but I think this camera will also appeal to some events photographers and so I will try to cover that too.
You can read about my initial thoughts on the X100F here.
Also see this detailed look at lens quality on the X100F.
….And also a direct head to head between the 24MP X-Trans III and a 24MP Full-Frame sensor in terms of sharpness and fine detail here.