It has been about 5 months since my last post, primarily because I've been very focused on my day job. Fret not though! I did get a chance to do some traveling recently, and have been spending quite a lot of time with the Sony A6000. I've used it while camping, hiking, and mountain biking at Lake Tahoe this past summer, it has been to India with me, and I've spent a bit of time shooting some landscapes with it here in San Francisco. Thus, I finally feel confident enough to write up a complete review of my experiences with it. If you can't already tell from the headline, I'm very happy with it, but there are several nit-picks that are worth digging into.Read More
The Fuji X-series of cameras have been some of the most discussed and reviewed cameras of the last two years. Fuji appear to have created a very successful formula that excites both amateur a professional photographers alike: good image quality in a small, reasonably-priced package that has intuitive manual controls. Combine this with a great lens lineup and a solid history of adding features via firmware updates, and one can see why they have become so popular. It's no secret that I shoot primarily with Nikon gear, but I'm certainly not the type to become attached to a particular brand when a better tool for the job exists. Since my Nikon setup is rather large and ungainly to travel with, I decided to give the Fuji X-E2 a try to see if it has what I need in a travel camera.
Before we begin, a quick note: I tested the X-E2 with the recently-released 2.0 firmware update which added significant electronic viewfinder (EVF) and autofocus (AF) improvements. Additionally, this is not intended to be an exhaustive review, but more of a short collection of my thoughts about the camera.
The X-E2 is a handsome little thing: the perfect size and weight for travel, and with the MHG-XE accessory grip, the camera fits perfectly in my hand, and has a solid, well-built feel. Fuji did very well with the control layout, adding manual switches wherever possible instead of slow-to-access menu items. I'm a big fan of the manual shutter speed dial up top, the AF-mode selection switch, and the aperture selection dial that is on most Fuji lenses. The software menus are fairly well-thought-out; I was able to get used to them very quickly, so no complaints there. This was the first camera with an EVF that I've used for any extended period of time, and I was pleasantly surprised by the EVF resolution: it's very good! The EVF refresh rate is also rather good (the 2.0 firmware update greatly improved this) - there's a barely-noticeable delay between what happens in real-life, and what you see in the EVF.
Unfortunately, that's where my praise for the handling aspects must end. The camera is very slow to start-up, and certain functions can't be accessed until a few seconds after startup. The ISO button, for example, doesn't work until 2-3 seconds after turning the camera on, which resulted in one missed shot for me. If the EVF is in eye-sensor mode, it's very slow to activate shortly after turning the camera on. The process would often go something like this for me: turn the camera on whilst bringing the EVF to my eye, and then wait 3-4 seconds for the EVF to activate. To me, this delay was unacceptable. The default EVF mode switches between the EVF and rear LCD based on an eye sensor, but three other modes are available: EVF only (always on), rear LCD only, or EVF only activated by the eye sensor. I think the camera could use one more mode: DLSR emulation, where the EVF is used to compose and take photos, and the rear LCD is used for photo playback and menu access (and remains off otherwise). The addition of this new mode, along with much faster EVF response/activation time after startup, would make the camera much more usable, and would eliminate the sluggish feel that it currently has.
The AF is fast and accurate in good light. It's slightly slower, and noticeably less-accurate in poor light. The 2.0 firmware greatly improved AF performance, but it's certainly not up to DSLR standards yet, and it's not even close to DSLR standards when tracking moving subjects. Despite this, it's quite good for a mirrorless camera, and certainly usable for a travel camera.
I will go ahead and say it up front: the X-E2 image quality is very disappointing. The X-trans sensor has good noise characteristics and delivers great dynamic range in RAW, but unfortunately this comes with a great loss of detail. Some of this detail loss is due to the RAW de-mosaicing algorithms in Lightroom (I'm using 5.4), which seem to give images a strange, painting-like rendering. However, some detail loss just seems to be a characteristic of the strange pixel arrangement that the X-trans sensor has, since some artifacts and detail-loss are still visible in the JPEGs produced in-camera. The strange, painterly effect is most noticeable when taking photos of fine detail, like vegetation, and ultimately is unacceptable for a landscape photographer like myself. Some photographers seem to be getting better results by converting RAW files using CaptureOne, but I'm stubborn and refuse to completely change my workflow to make minimal image quality gains. Ultimately, I'd be much happier with a traditional bayer arrangement; noise is always correctable in post-processing, but lost detail is never recoverable.
To see an example of this strange X-trans rendering, click through the image below to view a full-sized version:
In the image below, you can see the X-trans de-mosaicing struggle with the fine detail of the bridge cables. There is also a strange color-cast around some of the cables, which looks a bit like chromatic aberration, but is not removed by Lightroom 5.4's correction algorithms. Click through to see the full-sized image:
Another with some of the strange painting-like rendering in the details of the boats. Despite this, I'm fairly happy with this image. Click through below to see the full-sized image:
There's quite a lot of detail-loss in the long-exposure below. Click through to see the full-sized image:
The next image, I'm very happy with, and would consider quite good. For some reason, the camera seems to fare better when taking portraits. Click through below to see the full-sized image:
The X-E2 is a pretty good little camera, albeit with a number of annoying handling nags. I would be willing to forgive some of the handling problems if the image quality were up-to-par for my landscape work, but ultimately the X-trans sensor is unacceptable for me. Sadly, my landscape work has turned me into an obsessive pixel-peeper: I demand good detail from my RAW files, and the X-E2 doesn't deliver in that regard. Even the JPEGs leave a lot to be desired. If you're as landscape-obsessed as I am, I don't recommend this camera for you. My Nikon D7000 from 2010, which also has a 16mp APS-C sensor, produces RAW files with much more detail than the X-E2. If, however, you're a more casual photographer and not as pixel-obsessed as I am, then you could be very happy with the X-E2. It delivers outstanding web-sized images, and could make a great travel companion for you!
If Fujifilm is listening, here's what I'd like from my next X-camera:
- Replace the X-trans sensor with one that has a traditional bayer pixel arrangement, like the excellent 24mp APS-C sensor used in Nikon and Sony cameras.
- Address some of the camera's slowness and handling issues, particularly the slow responsiveness just after turning it on.
- Add a DLSR EVF mode: compose and take photos using the EVF, playback and menu access using the rear LCD.
Those three fixes would make the X-system a perfect travel setup for me. Until then, I'm going to stay away from Fuji cameras, whilst yearning wistfully at their beautiful prime lens lineup and compact-size.
- Size and weight: perfect for travel!
- Manual control layout: I really like the manual shutter dial up top, and how the majority of lenses come with an aperture ring.
- Fuji's lens lineup is outstanding, and is aimed perfectly towards enthusiasts and pros. Canon and Nikon could certainly take a hint here and add a number of fast primes to their APS-C lineups.
- Excellent dynamic range and noise performance.
- Produces beautiful web-sized images.
- X-trans image quality leaves a lot to be desired: lots of detail loss due to X-trans de-mosaicing.
- The EVF is slow to activate after turning the camera on: it takes 2-3 seconds for it to recognize that my eye is at the viewfinder after switching the camera on. This is incredibly frustrating, and makes the camera feel very sluggish.
- The EVF needs one more mode: "DSLR emulation", where the EVF is used to compose and take photos, and the rear LCD is used for photo playback and menu access (and remains off otherwise).
- The EVF doesn't adjust brightness automatically based on ambient light, and it's a bit too dark in very bright sunlight (even at its brightest setting).
- The ISO menu can't be accessed for a few seconds after turning the camera on, which sometimes results in a missed shot.
- IS.2, the secondary image stabilization mode, doesn't activate on shutter 1/2 press as described in the camera manual. Instead, it seems to activate during photo capture, which sometimes results in blurry photos since stabilization can take a short moment to properly "settle". IS.1, the default mode, leaves stabilization active constantly as long as the camera is on, but unfortunately this drains the battery much faster.
- Spot metering doesn't use the selected AF point, and instead the center area is always utilized. This is very counter-intuitive to how every other camera meters in spot mode.
- Long exposures cannot be stopped mid-exposure - not even by physically turning the power switch off. This is rather frustrating for a landscape photographer, since sometimes I start a long exposure by accident and need to end it quickly to change a setting and re-start.
- Long exposure noise-reduction is a mystery. It doesn't actually take a second dark-frame exposure to subtract noise from the original.
- AF isn't up to DSLR standards, particularly in low-light.
- Exposure delay/self-timer mode not preserved when turning the camera off and on. I use the 2-second self-timer as an exposure delay for landscape photos, so I have to re-enable it each time after turning the camera back on.
- Unable to change selected AF point without first pressing a different button. This extra step is an annoyance coming from a DSLR.
These two prime lenses have similar focal-lengths, similar maximum apertures, and similar price-points. It's no wonder that I see them discussed and compared with each other on forums fairly often. Given $700-900 to spend on a fast, relatively-wide, prime lens, which do you get? First, let's take a look at a high-level specifications summary to get a feel for what they offer (clear winners highlighted in green):
To summarize: the Nikon is a bit wider, smaller, and lighter than the Sigma. It's also a bit slower (2/3rds of a stop), but it's a good deal less expensive as well.
Construction and Handling
The Sigma really outshines the Nikon in both construction and handling. When I first got my Nikon 28/1.8G, I was disappointed with the light, plastic feel of the lens. The focus ring feel light and cheap. It most certainly doesn't feel like a $700 prime lens should. Unlike the Nikon, Sigma uses a metal lens barrel with a very smooth, well-damped focus ring. It's also larger and heavier than the Nikon, which really give the Sigma a nice, high-quality feel. The flip-side of that is that the light weight and smaller size of the Nikon can certainly work to its advantage. I prefer to travel light, especially when hiking, where every single ounce really adds up. In that sort of situation, the Nikon would be my go-to lens.
Focal lengths are tricky. What works perfectly for one photographer may feel completely wrong for another. To that effect, I've created two GIFs, below, which help to illustrate the difference in focal lengths. Please note that these comparisons are for the field-of-view on full-frame / FX cameras.
As you can see, the fields of view these lenses provide are actually quite similar. The Nikon may be slightly better suited towards landscape photos, and the Sigma might be a bit better suited for things like portraits and street photogoraphy. However, I like to challenge these sorts of assumptions, and I really think that either lens could be excellent for landscape, street, travel, and most other general types of photography - when placed in the right hands, of course. To give you an example, below is a landscape photo taken with the Sigma 35/1.4, and an environmental portrait taken with the Nikon 28/1.8.
I don't have the right resources or knowledge to do a full optical assessment of both lenses. The general consensus across the web is that the Sigma is the new resolution benchmark to beat. From what I've seen, I have to agree. It's pretty astonishing in terms of center resolution and sharpness.
That said, the Nikon is nearly as good as the Sigma, and I have absolutely no complaints about its resolution or performance. If you're eager to look at some charts and numbers, the brilliant folks at Photozone.de have reviewed both lenes:
As you can see, the resolution numbers are quite similar. In terms of flare-resistance, both lenses performed very, very well during testing. I didn't have any issues with flare, even with the sun in the frame. It appears that Sigma's multi-coatings are beginning to rival Nikon's expensive "Nano-Crystal" coatings.
In short: both lenses are optically excellent.
I enjoy bokeh (out-of-focus blur) as much as the next guy, but I tend not to obsess over it. In this case, the Sigma wins out over the Nikon 28mm f/1.8G. This is simply due to the slightly longer focal-length and faster aperture, which just makes it more "bokeh-rific" than the Nikon (given the same subject at the same distance). Nevertheless, the Nikon's bokeh is pleasing and soft in certain situations, and about as strong as one might expect from any 28/1.8.
Both lenses feature fast, quiet, ultrasonic motors. I've encountered reports of the Nikon 28/1.8 having some focus-shift issues at different apertures (in-focus at f/1.8 may be out-of-focus at f/5.6). I've thoroughly tested my copy, and have had no such problems. There is some noticeable field curvature that one must be aware of, but that's expected of a moderately wide, fast lens. I didn't notice any field curvature with the Sigma.
As far as accuracy is concerned, both lenses are very accurate. Over the course of my week shooting with the Sigma, I only experienced one hiccup and missed-shot due to an AF problem. This was while using the extreme, center-bottom focus point. Some users have reported problems with the Sigma on the D800 when using the extreme outer focus points, but other than my one missed shot, I didn't have any issues. In fact, I also missed a shot in a similar situation with the Nikon during the same time period. Overall, I'd say that these lenses are about equal when it comes to AF performance and accuracy. Neither one is absolutely perfect, but they're not terrible either. In fact, they're both quite good.
I've been taking photos with the Nikon 28/1.8G for nearly a year now, so I've had time to amass a number of sample images that I'm very happy with. Unfortunately I only had the Sigma 35/1.4 for less than a week, since I rented it from the wonderful people over at LensRentals.com. Nevertheless, I managed to get a handful of images that I'm content to display.
Summary & Conclusion
You really can't go wrong with either of these lenses. If cost, weight, and build quality are of negligible concern, then the decision could boil down to just one very subjective factor: focal-length. It really all comes down to the photographer's preference. Gary Winogrand famously used a 28mm lens for most of his street photography, but Henri Cartier-Bresson shot primarily with a 35mm.
For landscapes, I usually reach for a zoom (typically the 17-35/2.8 or 24-120/4) so that I can compose more precisely. For me, primes are more useful for travel, street, and portrait photography. Because of this, I find the slightly tighter focal length of the Sigma more appealing for general use. I also prefer the heavier and more solid construction of the Sigma over the Nikon. Conversely, I can see myself reaching for the Nikon when traveling light, like when hiking, where every ounce really counts. Other photographers might prefer working with primes for landscape photography, where the 28mm might be more appropriate than the 35mm.
If you're having a hard time deciding, I suggest renting these lenses from LensRentals.com before you choose. Or, you might try scotch-taping your zoom lens at each focal-length for a few days. It'll quickly become apparent which one you prefer.
Sigma 35/1.4 Pros
- outstanding optical quality
- 2/3rds of a stop faster than the Nikon
- robust construction
- nice, normal-wide field-of-view on FX
- on DX it yields a "normal" 50mm equivalent field-of-view
Sigma 35/1.4 Cons
- more expensive than the Nikon
- no rear dust gasket/seal on mount
Nikon 28/1.8 Pros
- outstanding optical quality
- light and relatively compact design
- nice, moderate-wide field-of-view on FX
- less expensive than the Sigma
- has rear dust gasket/seal on mout
Nikon 28/1.8 Cons
- 2/3rds of a stop slower than the Sigma
- cheap-feeling construction
- on DX it yields a slightly strange 42mm field-of-view