Fujifilm 18-135mm f/3.5-5.6 Review

bythom fujifilm 18-135mm.jpg

What is It?

Let’s be clear up front: I’m not a fan of superzooms. At 18-135mm (28-200mm equivalent), this lens qualifies as the do-everything superzoom type. With most of these superzooms there are always tradeoffs being made. Two that are clear right up front are that this lens doesn’t go particularly wide (24mm equivalent would be far better), and the maximum aperture isn’t particularly fast.

Let’s deal with that second part first:

  • 18mm = f/3.5
  • 23mm = f/4
  • 35mm = f/4.3
  • 55mm = f/5
  • 70mm = f/5.3
  • 135mm = f/5.6

A fairly linear degrading of maximum aperture, unlike some other designs I’ve seen that get slow very fast. Still, in any real focal length other than it’s modest wide angle ability, we can’t really call this anything but a slow lens.

This zoom is also a reasonably large lens, almost 4” in length and 3” wide at the front element (67mm filters). The lens extends considerably at 135mm (almost 6” without hood), and the front elements rotate slightly during zoom. At 17.3 ounces (490g), it’s also not light. Some of that is the use of a lot of metal in the construction, as usual for Fujifilm, but 16 elements in 12 groups is also a fair amount of glass, too. Four of those elements are aspherical, two anomalous dispersion.

As usual with Fujifilm, you also get a thin “bag” for the lens and a bayonet lens hood.

Fujifilm claims the OIS in the lens can correct up to five stops, which is about state-of-the-art for stabilization systems.

The lens body is mostly metal, the lens hood plastic. The lens is made in China.

How’s it Handle?

In terms of handling, Fujifilm has made a choice that I still find problematic. When the X series first kicked off, all the lenses had aperture rings. Then two things happened: first, Fujifilm came up with some variable aperture zooms; second, Fujifilm dipped into the low-end body range with cameras that didn’t exactly have the direct controls necessary to substitute for an aperture ring. The result was a ring that controls aperture that is click stopped, but not marked, and a switch on the side of the lens to set auto aperture versus user control of apertures.

I’m not a fan of that approach. And here we have yet another lens using that construct.

The aperture “ring” (near the back of the lens) is too easily moved from its click stops, a common problem amongst the lenses that share this design, and my 18-135mm was one of the easiest to move the ring on. Since I shoot in Aperture priority mode most of the time and Manual exposure mode most of the rest of the time, I’ve found that I’m having to watch that I haven’t accidentally changed apertures. That’s a small thing, but it’s a nuisance when the aperture gets changed and I don’t notice right away.

The front element of the lens moves with zooming (but not focus), something to note if you’re a fan of polarizing filters. Fujifilm also doesn’t seem overly consistent in their filter sizes in their lens lineup. This lens is 67mm, which is shared with only a couple of other lenses (the 16mm f.1.4, for example).

Heft of the lens is substantial. On the lighter X bodies even this modest sized lens is going to create a bit front-heavy of a system. On a X-T1 or X-T10, this is still heavy enough that you'll want to put your hand under it for support while shooting.

The zoom ring is a bit stiff on my sample, while the focus ring was somewhat loose. Still, both were relatively smooth and I was able to use them without overshooting. While the focus ring can be differentiated from the zoom ring by feel (zoom ring has a more rubbery touch), I wish there was a little more distinction here.

The supplied petal lens hood bayonets onto the lens and can be reversed on the lens for transport, though it makes the front of the lens even wider when you do that. The front pinch cap has such wide pinch points that you can’t miss them, as you can on some lenses. Thank you Fujifilm.

Overall, the build quality is what you expect, with attention to detail, great materials, and tight fits all around.

For US$900 you expect a good lens. In terms of build and handling, the 18-135mm certainly seems up to the level I expected, except for that aperture ring function I don’t like.

How’s it Perform?

I wish I could say the 18-135mm was a top performer, but as with most superzooms, it’s not. The central area is generally very good, with edges also being very good at the marked 18mm and 23mm positions. Unfortunately, above that I saw a number of issues. First, it appears my sample was miscentered a bit as it zooms. At 55mm the sharpest area had shifted left enough so that the left and right sides of the frame were slightly different in terms of acuity. I’ve seen other tests of this lens that seem to indicate something similar, so I have to think that the cam positioning elements during the zoom isn’t doing it’s job of holding them perfectly aligned to the mount (remember that slight front element turn during zoom? I’ll bet that’s the culprit).

Still, in terms of acuity and resolution, this isn’t a bad lens. The side-to-side variations I saw were somewhat minor (almost ignorable). The corners are never terrible wide open, though left and right sides might be slightly different. At f/8 and f/11 the lens was very well behaved and about as sharp as I expected it to be. Any centering/linearity issues were totally muted in the results at those apertures and not problematic. Still, this isn’t nearly the sharpest lens Fujifilm has made. It’s a superzoom that has a lot of compromises in it and thus performs decently, not great.

US NV Las-Vegas Aug-2015 XT10 43939.jpg

Images like this one reveal some of that corner issue, though it may be difficult to see at this size. The Eiffel Tower and hotel in the background are still within acceptable DOF limits, at least in their mid areas. As I look closer and closer into the very corners, I can see the corner sharpness definitely degrade, with the absolute corners being marginal at best, especially on the right side of my sample. Still, the primary portions of the frame where your subject is likely to be render just fine.

Curiously, vignetting and linear distortion were actually pretty well handled in this lens, even in raw files with uncorrected data. Indeed, so good that I’d be tempted to not correct the data (that means far less than a stop vignetting and .5% or less linear distortion pretty much across the board).

There’s no such thing as a free lunch, so the thing that probablydoesneed correcting is chromatic aberration. While not terrible, it’s persistent and worse in the lower middle of the focal length range.

One thing that wasn’t quite as good as I expected was the OIS. It’s good, but it’s not five stops good in my estimation. I had to watch my lower shutter speed boundary fairly careful or else I got a somewhat weird single axis blurring:

US NV Las-Vegas Aug-2015 XT10 43934.jpg

Vertical lines are tight, horizontal information seems doubled and overlapped.

I’d judge the OIS to be effective to maybe three stops, and even then I could see some minor axis-related movement.

Examining images from the lens has me saying that many—maybe even most—of you will be satisfied with the lens. Other than the miscentering I saw in my sample there aren’t any glaring faults, just anumber of limits you want to avoid, if possible. It’s a superzoom with lots of compromises that attempt to make it useful for many purposes. This is not a lens I’d select to optimize my image quality, but it is one that I might consider as an all-purpose walk around lens.

Final Words

Another lens from Fujifilm where I was hoping it would perform better than it did. A reasonable and decent performer, but not exceptional in any way that I can see. The miscentering disturbs me, though I had to look hard to find the results of that in my image files. As an all purpose walk around lens without any exceptional traits, it feels a bit overpriced to me, but I note that you can find it for less than the original list price, so look around before purchasing.

Support this site by purchasing from this advertiser:

text and images 2017Thom Hogan
portions Copyright 1999-2016 Thom Hogan-- All Rights Reserved
Follow us on Twitter: @bythom, hashtags #bythom, #sansmirror