Fujifilm X-M1 Digital Camera Review

Much like the Panasonic GF6, Olympus E-PM2, and Sony NEX-3N, the X-M1 offers a mashup of point-and-shoot handling and top-flight image quality, attempting to serve as a bridge for compact camera users transitioning to system cameras. Sure, Fuji's effort gives shooters a bit more to work with, but it has a correspondingly higher price tag. While Sony and the Micro Four Thirds manufacturers are working to push entry-level mirrorless prices below the $500 threshold, the X-M1 will run you closer to $1,000.

But while Fuji's junior X may have shed much of the X-series mystique, it retains one of its predecessors' most important features: the 16.3-megapixel X-Trans sensor. It's a chip that produced stellar results in the X-E1 and X-Pro1, and being able to get it at a lower price is likely the best reason to consider the X-M1 over its rivals. Read on to find out how it fared in our lab and real-world testing.

Design & Handling

A victory of style over substance

We often advise that you should try to get a camera in your hands before you buy it, and that's doubly true for the X-M1. This baby X may look a lot like its big brothers at first glance, but it's a total con job.

The stylish, retro-inspired body is absolutely gorgeous, but one touch of the top plate is enough to shatter the illusion—chintzy, hollow plastic destroys any pretense of luxury.

The material covering the smallish front grip isn't particularly comfortable, and the dials vary drastically in quality. The rear thumb rest is hard plastic with a few raised bumps. It isn't the slippery soap bar we've bemoaned in cameras like the Nikon 1 J3, but the X-M1 is far from the most accommodating design we've encountered. Overall, it's a huge disappointment, and one that contributes to a user experience we've come to associate with cameras costing hundreds of dollars less.

{{gallery "product"}}

On the top plate, the dedicated shutter speed and exposure compensation dials from earlier X-series cameras have been replaced with a single, context-sensitive wheel and a physical mode dial. The customizable function button remains, providing you one-touch access to any of 15 different vital settings. A secondary control dial is positioned awkwardly above the thumb rest, facing up. Beyond its cheap feel and uncomfortable placement, it's also awfully imprecise in use. While there are click stops, the resistance is so light that we often found ourselves overshooting the shutter speed, aperture, or ISO setting we wanted.

Below the thumb rest is a standard d-pad control cluster. The four directional buttons provide direct shortcuts to AF point selection, macro focusing, drive modes, and white balance options. Above are playback and video recording toggles, while below are display and quick menu buttons. As you'd expect, the bulk of the backside is taken up by the 3-inch LCD, which tilts up a little more than 90 degrees, and down nearly a full 90 degrees. That last spec is pretty impressive, letting you take overhead shots at parties and concerts with ease. The screen is bright and clear, but its refresh rate slows down quite a bit in dim light.

The X-M1 ships with a new Fujinon XC 16-50mm f/3.5-5.6 OIS kit lens. The "XC" designation indicates that it's part of Fuji's newly established line of low-cost lenses, as opposed to the fancier XF models. The main differences? No aperture ring and a plastic lens mount, along with a more conservative optical design. Despite its cheaper construction, the new kit handles quite well, and our tests showed it to be a capable performer (more on that later). Unlike many kit lenses, the front element doesn't rotate as you zoom, which means you can use filters without worry.

{{ gallery "tour" }}

Features

More for newbies, less for vets

Where the previous X-series cameras were aimed almost exclusively at veteran photographers, the X-M1's mission is to spread the love to a less experienced audience. As such, the physical mode dial is host to a mess of automatic shooting modes, scene selections, and picture effects—niceties the X-E1 and X-Pro1 turned their noses up at.

In addition to the regular Auto mode, there's an SR+ Auto option that takes absolute control of the X-M1's toolbox to auto-selects from among 58 scene presets. The SP setting lets you manually choose from 10 of those, while Adv. accesses multiple exposure and an array of 13 creative filters. Five of them are selective color modes, but you'll also find popular options like Toy Camera, Miniature, Pop Color, and High/Low Key.

As with all other X-series models, you can take advantage of several Film Simulations that attempt to replicate Fujifilm's legendary film stocks. The X-M1 doesn't have as many options as the X-E1 and X-Pro1—in particular, it's missing the Pro Neg settings. But it has Provia (Standard), Velvia (Vivid), and Astia (Soft), along with black-and-white and sepia.

The 16-50mm kit lens is an outstanding feature in its own right. The build might be sub-optimal, but the 24mm effective wide-angle is a wonderful bonus that gets you an extra-wide perspective compared to the average 18-55mm kit. It's extremely useful for landscape and architectural shooting, where the tilting LCD is another excellent addition.

The lens is also optically stabilized; Fujifilm claims the OSS technology will get you 4-4.5 stops of shake-resistance, and our subjective experience seems to bear it out. In informal testing, we were able to get sharp handheld shots with shutter speeds as low as 1/5sec at 16mm and 1/10sec at 50mm. Not bad!

Video isn't a priority for Fuji, but you still get the option of 1080p or 720p recording, both at 30fps. Manual control is limited, but you can choose the aperture, shutter speed, film simulation, and exposure compensation before you begin recording—once you start rolling, it's all locked down. There's continuous autofocus during video, but it's slow and tends to hunt for a focus lock. On the whole, we recommend focusing manually or using a hyperfocal setting.

Lastly, the X-M1 offers embedded WiFi connectivity. The functionality is simple and straightforward: you can transfer photos to a phone (or tablet), browse images on the camera from your phone, or geotag photos on the camera using your phone's GPS. Unfortunately, the app crashed incessantly on my rooted Galaxy Nexus running Android 4.2.2, so I couldn't try it out for myself. Unlike competing WiFi implementations, the Fujifilm Camera Application doesn't let you use your smart device as a viewfinder or control the camera by remote. It's a missed opportunity, and something we hope they'll improve upon.

Performance

It's hard to argue with these results.

Regardless of any issues it may have with fit and finish, handling, and controls, the X-M1 has the ability to produce absolutely gorgeous photos. Details are crisp and colors are vivid (if not clinically accurate), creating shots that have a natural power to hold the eye. As with other X-series cameras, we had great fun playing with the various film simulations—each produced striking results.

{{gallery "sample-photo"}}

In the lab, the venerable X-Trans sensor and new kit lens proved their mettle. Resolution figures in our sharpness test were sky high throughout the zoom range. As usual with a consumer-grade lens, center sharpness was much higher than the edges and corners, particularly shooting at maximum aperture. The JPEG processing is tastefully done—edge enhancement doesn't produce notable haloing, and noise reduction is fairly subtle.

Regardless of the noise reduction setting we selected, noise stayed below 2% up through ISO 6400. The X-M1 employs a five-stop NR scale, ranging from -2 to +2 (0 is selected by default). Going to -2 resulted in a loss of about a stop of noise performance, and going to +2 resulted in a corresponding gain. The two highest sensitivities are clearly pushing the limits of signal amplification—showing exposure shift, saturation loss, and a drop in contrast—but they can be used to good effect in black and white.

We encountered the same dynamic range issues with the X-M1 that we saw with the Fuji X100s. The camera's JPEG engine aggressively clips blacks to keep noise down in the shadows. That means you need to shoot RAW if you want to recover dark areas in post. Don't get us wrong: The X-M1 produces beautiful JPEGs. You just need to make sure they're properly exposed, because you don't have as much latitude to push or pull as you might like.

Pixel-perfect color isn't Fuji's aim, so it's no surprise that the X-M1 didn't ace our color error test. The camera's most accurate color mode (Provia) returned a saturation-corrected color error of 3.02, while a score of about 2.5 is more typical of most contemporary system cameras. Saturation was much higher than ideal in all color modes, hitting 106.8% in Provia and a whopping 135.2% in Velvia.

Video performance was awful. Movies recorded in 1080/30p were sharp, but suffered from excessive moire, aliasing, and artifacting, even in good light. Sharpness dropped off a little in dim light and artifacting became quite obvious. Low-light sensitivity was surprisingly poor, as the X-M1 needed 22 lux of ambient illumination to achieve the target 50 IRE on a waveform monitor.

For more detailed test results, visit the Science Page of this review.

Conclusion

Et tu, Fuji?

At first glance, the Fujifilm X-M1 seems like a commonsense proposition: Take the retro styling and superior image quality of the prestigious X Series and bring it down to a more consumer-friendly price point. Why wouldn't Fuji want to open up a new user base and reap the profits, after all?

But if you think about it for a moment, a few problems come into sharp focus:

1. Fuji has minimal brand recognition with contemporary consumers.
2. An $800 price tag is not consumer-friendly in today's system camera market.
3. The X-M1's cheap construction doesn't match up to looks or price, undermining its credibility.

The first point is a bit of a catch-22—if companies don't attempt to court beginners, they'll never gain any ground in the market. But the Fujifilm's recent resurgence has been entirely driven by enthusiast adoration. We're just not sure it will pay dividends for such a small company to divide its focus and dilute its appeal.

The X-M1's cost is a more serious issue. Yes, $800 with an above-average kit lens seems like a surprisingly low asking price for a Fuji X, but only until you actually hold this thing. It feels more like a $600 or (given the existence of the Sony NEX-3N) $500 camera. Granted, the X-Trans sensor produces more attractive results than the NEX-3N can offer, but $300 better? It's your money, so you'll have to be the judge on that score.

Putting an X-Trans sensor in a plastic body is a bit like putting a Ferrari engine in a Toyota Camry. Yeah it's got a lot of get up and go, but at the end of the day, the beige paint and cloth seats diminish the overall experience. And the handling? Not so sporty.

If you're willing to spend $800 on a Fuji X-series camera, chances are you're the kind of shopper who's looking for ultimate quality, both in image quality and build. So do yourself a favor: Spend a little more and pick up an X-E1. For the extra coin, you'll get improved controls, the confidence of an all-metal build, and the same great image quality.

And if your budget can't stretch beyond $800, there are still plenty of quality options available to you. We loved the Sony NEX-6, which offers similar image quality, better controls, and an electronic viewfinder, all in a sturdier body. If you don't mind a little extra bulk, the Pentax K-50 gets you a 100% coverage optical finder, a reliably awesome sensor, magnesium alloy build quality, and weather sealing—all for less than the X-M1.

And that's just the tip of an iceberg that's also home to the Olympus E-PL5, Ricoh GR, Nikon D5200, and many others. After all, if there's one thing that the X-M1 has taught us, it's that you need to look below the surface.

By the Numbers

We have plenty of qualms about the Fujifilm X-M1's build and features, but we couldn't find much wrong with its performance. In our lab testing, the X-M1 had a performance profile neatly in line with previous X-Trans cameras, like the X-E1 and X100s. Sharpness was spectacular, even with the cheaper Fujinon XC 16-50mm kit lens. The JPEG engine displayed a gentle hand with edge enhancement and noise reduction, and dynamic range was quite good in RAW (while showing some odd behavior in JPEG). Color accuracy is the only real disappointment, but that's never been Fuji's goal, anyway.

Resolution

Every single Fuji X-series camera has produced off-the-charts sharpness, from the fixed-lens X100s, to the X-Pro1 with the 35mm f/1.4 lens, to the X-E1 with the 18-55mm f/2.8-4. The X-M1 is no exception, paired with the new XC 16-50mm f/3.5-5.6. Despite its more pedestrian build and maximum aperture, the new lens is still quite the performer—we're tempted to say it's the best consumer-grade kit zoom we've encountered.

Sharpness wide open is at least very good at all focal lengths. The sharpest focal length is 16mm, but full wide angle also shows the most resolution falloff toward the edges and corners. Shots at 35mm and 50mm were less sharp overall, but more consistent across the frame. Stopping down to f/8 or f/9 produced the highest resolution figures across the board, which is absolutely typical for an APS-C sensor and lens.

Noise & Noise Reduction

The X-M1 produces more noise on the whole than other X-Trans equipped cameras we've tested, losing about a stop of noise performance to the X-E1 and X100s. With noise reduction set to its default setting (0), noise starts at 0.56% at ISO 100, hits 1% at ISO 1600, and nears 2% at ISO 12800. Cut the NR by two clicks and you'll lose a stop of noise; bump it up by two and you'll gain a stop.

Of course, adding that much noise reduction sucks away the detail. Smoothed JPEGs look pretty good, even with the highest level of NR enabled, but only when reduced to web size. When viewed at 100%, the loss of detail is difficult to miss. We found that shooting monochrome JPEGs with NR turned all the way down added a nice filmic grain, giving shots a gritty feel. Your mileage may vary.

Dynamic Range

Looking at shots straight out of the camera, you wouldn't think the X-M1 has any issues with dynamic range. Shadows and highlights can coexist quite comfortably in the same image, and JPEGs look beautiful in the various film simulation modes.

But if you significantly underexpose your JPEGs, you're going to run into trouble in post. Even at the lowest ISO sensitivities, the processing engine reduces all shadows to pure black. That produces punchy, high-contrast shots without visible shadow noise, but if you need to recover any shadow detail you simply won't be able to. Our advice? Shoot RAW, or JPEG + RAW, and you'll always have recourse if you let yourself down.

Color & White Balance

One area where the X-M1 significantly underperformed the other X-series cameras was color accuracy. Why? One simple reason: no Pro Neg. film simulations.

Pro Neg standard has always been the top-performing color mode on previous X-series cameras, and it's been cut out of the X-M1. As a result, Provia (the default) is the most accurate, with a ∆C 00 corrected color error of 3.02. To put that number in perspective, the best score we've ever seen came from the $6,800 Canon 1D X, which returned a 1.7. The X-Pro1 scored a very respectable 1.99, and most modern DSLRs come in at 2.5 or better.

It's not surprising that a consumer-oriented camera produces punchier, more vibrant shots, but we're not sure we see the sense in removing more accurate film simulations when it costs Fuji absolutely nothing to include them.

White balance, on the other hand, was the best we've ever seen from an X-series camera. Automatic white balance was on par with the mighty X-Pro1 and significantly outstripped the X-E1. As usual, incandescents provided the most difficult challenge, resulting in an average color temperature error of 1645 K; compact white fluorescent (515 K) and simulated daylight (75 K) proved far more accurate, and can be enjoyed without fear. Custom white balance was spectacularly accurate—daylight was actually the worst, with an error of 103 K, while incandescents and compact white fluorescent were virtually perfect at around 40 K.