Anybody who’s used old school manual focus lenses knows what a lens should feel like. There’s a feeling of density and smoothness that’s sadly disappeared form modern lenses. Gone are knurled metal lens control rings, and thick metal lens barrels, in their place we have rubber and plastic. Which brings me to the EF-M 18-55mm f/3.5-5.6 IS STM.

Through and through the build is quite nice for a $250 list kit lens, though admittedly, it’s also one of the more expensive 18-55s on the market, or, for that matter, general purposes zoom kit lenses for mirrorless systems, as well. The gunmetal gray barrel is unmarred by switches, bulges, and rubber grips, giving it a certain seductive streamlined look to the lens that invokes a feel of Leica M or Zeiss ZE lens than a kit lens for an entry level mirrorless system or even your typical L zoom.

The external surface of the lens is divided into essentially 3 zones, from front to back, the focus ring, the zoom ring, and a section of fixed barrel in front of the lens mount. The layout of the controls is somewhat interesting. It follows the typical kit lens scheme, with a smaller focus ring at the very front of the lens followed by a larger zoom ring—in contrast to the mid-tier layout of a large zoom ring ahead of the small focus ring, or the pro layout with a large focus ring ahead of an equally sized zoom ring.

I’ve read many reviews that imply the focus ring is a tiny nub similar to that of the EF-S 18-55mm f/3.5-5.6 IS STM and its predecessors. This isn’t strictly speaking accurate. The focus ring is actually 9/16^ths of an inch wide, stretching from the 1/8^th inch wide knurled portion all the way back to the seam between it and the zoom ring. In fact, the entire rotating portion of the ring is actually wider than the ring on an EF-S 10-22mm f/3.5-4.5 USM or EF 28-135mm f/3.5-5.6 IS USM.

Like all stepper motor (STM) lenses, the EF-M 18-55mm f/3.5-5.6 IS STM is a focus-by-wire design, meaning that there isn’t a physical connection between the focus ring and the helical that moves the lens elements.

The focus-by-wire design means there are no stops on the focus ring; not even soft stops like you have on a lens with a ring ultrasonic motor. The focus ring turns freely, and smoothly, around the entire 360° in either direction, indefinitely. Admittedly, the feel of the ring is really nice and it very much reminds me of a manual focus lens in how it feels.

On the other hand, it also means that the operation of the lens is entirely dependent on the camera it’s mounted on, in this case the EOS-M. Let me start with the basics, if the camera is off, the lens won’t focus at all. With the camera on the way the lens behaves depends entirely on how the camera is setup.

Unfortunately, this is one place where the super clean design of the lens can be a bit annoying. There’s no auto-/manual-focus switch on the lens, and there’s no physical AF/MF switch on the the current EOS-M to make switching between auto- and manual-focus modes qucik.

The current EOS-M provides 3 focus modes you can select though the shooting 2 menu (red camera with 2 dots).

First, AF only. In this mode, then the lens’s focusing ring is disabled. No amount of turning it will affect the lens’s focus.

The second option is full manual focus mode, which works like any other manual focus lens. At any time the camera is on, turning the focus ring will cause the lens to focus.

The third, and final, option is AF+MF. In this mode, the lens’s focus ring will adjust focus only after the AF system has tried to focus (either achieving focus or giving up) and while the shutter button is being held in the half pressed. Strictly speaking, all Canon focus-by-wire lenses behave this way including all the STM lenses in EF and EF-S mounts.

Once you’ve used it a couple of times, the focus by wire system isn’t entirely bad. I’ve never noticed any significant focus lag, however, without a distance scale or hard stops it’s hard to tell. Likewise, it’s impossible to tell if the lens is focused to infinity or close focus simply because there is no stop or distance scale, and no indication on the camera that you’ve reached the stop.

Since there’s no need to directly couple the focus ring and the helical that positions the lens elements, the lens can use an inner focusing system without the added cost of designing in a way for the two components to be linked. This in turn means the front element neither rotates nor moves when the lens is being focused. A boon for people who wish to use creative filters—circular polarizers or split neutral density—on this lens.

The stepper motors do an amazing job at being quiet. If you thought ultrasonic motors were silent, the STM lenses have them beat.

Immediately behind the focus ring is the zoom ring. Like the focus ring, the full width of the zoom ring isn’t knurled, only the front 5/8” of the 1” wide zoom ring is knurled. The zoom ring rotates very smoothly throughout its 45°-throw covering the range from 18mm to 55mm.

The EF-M 18-55mm f/3.5-5.6 follows the standard zoom design where the inner lens barrel extends as the focal length increases. There’s only a single inner lens barrel, so the force needed to move the zoom ring remains constant regardless of the direction the camera is pointing or focal length. Zooming from 18mm to 55mm causes the lens barrel to extend 1-1/16^th inches.

The final, rear most, 5/8” of the lens’s outer barrel is fixed with the zoom scale indicator, mount position indicator and other miscellaneous lens information and branding printed on it. Behind that is the metal lens mount.

The big question though is how good are the optics. The short answer is that they’re good, not super-telephoto L prime good, but much better than I expected from a kit 18-55mm lens. Unfortunately, that’s about all I can say on the subject. I don’t have a test chart capable of exceeding 800 line pairs per picture height, and even if I did I don’t have a feasible way to test in a repeatable accurate setup. If you’re the numbers kind of guy, there’s always DXOMark, and SLRGear for that, and if you’re not so quantitative, there’s the Digital Picture’s lens test comparison charts.

Distortions seem to be fairly well controlled. There’s some noticeable barrel distortion at 18mm, that’s largely gone by 28mm. From 28mm to 55mm, while there’s technically some distortion, it’s minimal enough that I don’t see it in actual pictures.

The lens does a decent job at controlling flare, though it’s certainly susceptible to it. Like all non-L lenses, the lens doesn’t ship with a lens hood. For those wanting a hood to assist in controlling flare in some circumstances, the optional hood is the EW-54.

The hood situation with Canon lenses is a singularly frustrating point, they are the only camera company that refuses to put a hood in the box for every lens they sell—it’s a point reserved for the L lenses alone it seems. Fortunately, the EW-54 is not nearly as expensive as some other Canon hoods; it lists on B&H and Amazon at about $15. Going with a third party alternative, saves another $5. That said, Canon outrageously wants $30 for the hood though their online store.

The lens is optically stabilized, with what appears to be the traditional in lens stabilizing system (as opposed to they hybrid-IS system used in some Canon lenses that can correct for small pitch and yaw movements in addition to vertical and horizontal translation movements). Canon claims the stabilizer it’s able to provide shake reduction for 4 stops of shutter speed beyond that suggested by the hand holding rule of thumb. In practice there’s a bit of a problem with that assessment though.

The hand holding rule of thumb (that the minimum shutter speed needs to be 1/focal length or faster), is empirically derived from camera systems that used 3 points of contact to steady them, your face/head to the viewfinder, and both hands on the camera body and lens. The EOS-M adopts a usage factor similar to that of a point and shoot, where you hold the camera with both hands out in front of you with no 3^rd point of contact.

In general when using this camera, I don’t find you can simply apply the handholding rule of thumb. Moreover, the amount of instability in the way I hold the system seems to be much more inconsistent to what I see in an SLR. Some days I can get sharp pictures at the rule-of-thumb suggested shutter speeds, other days I can’t.

My point here is twofold; first, the image stabilizer is almost a necessity in my opinion. There’s just too much instability to not want to have that kind of shake protection. Secondly, the inherent instability of the platform makes testing the IS system’s capabilities more difficult—especially without some kind of controlled shaking test rig.

To sort of outline the difficulty of doing this, when I was trying to do an IS shake test for this article, I figured on a procedure as follows. With IS off, shoot a series of images with increasingly slow shutter speeds starting above the rule suggested minimum, and working down towards some point below the minimum. Then turn IS on and repeat the procedure.

In the processing of shooting this I found with IS off, I had blurred images at 1/160^th at 55mm, and sharp images at 1/40^th at 55mm in neither case with IS running. As I went back to reshoot another series, I found that having held the camera, and it’s by no means a heavy one, out in front of me for 5 minutes straight, my arms were tired enough that I couldn’t maintain any consistency.

For video shooters, the IS system operates continuously while recording. The dynamic IS does a very good job at removing the high frequency camera shake that is, in my opinion, the biggest problem with hand held footage.

Further adding some impressiveness to the IS system is the silence of it. Like the stepper motors, the IS system makes no perceptible noise when it starts or when it’s running. It’s by far the quietest IS system I’ve heard to date in a Canon lens, though admittedly most of my lenses are L lenses that were designed with no consideration for video work at all, so the slight whir as the IS system spun up wasn’t remotely relevant.

Finally, the switch free nature of the lens, means that the IS system is enabled or disabled though the camera’s menus. On the current EOS-M body, the camera warns you when the IS system is off by displaying an IS off icon, but not when it’s active. Further, the IS state of EF and EF-S lenses attached via the EF-EOS-M adapter do not indicate the state of there is system at all.

With all that said, the lens doesn’t come off scot free in my view. The range is decent, the image quality is good, but the f/3.5-5.6 aperture is sorely lacking to me. One has to understand, I’m coming at this lens and the EOS-M platform as a backup/fun body to my full frame EOS DSLRs and L zooms, so going back to an f/5.6 lens at angles-of-view where I’m normally shooting f/4 or f/2.8 isn’t an especially thrilling proposition.

I can certainly appreciate why Canon chose to the route they did, it didn’t do anything to differentiate the fledgling EOS-M platform form its competitors. I can’t help but wonder if the kit lens was an f/2.8-4 or even fixed f/4 if it wouldn’t have at least helped attract some more interest to the fledgling platform. That said, as halfhearted as Canon’s entry into the mirrorless market may have been, the EF-M 18-55mm f/3.5-5.6 IS STM isn’t half assed.

Lens Data