Carl Zeiss designed and Sony manufactured Sonnar T* 135mm f/1.8 ZA was one of the first Carl Zeiss lenses exclusively licensed for the (then new) Sony Alpha system. The lens was introduced under much fanfare in late 2006 and instantly became a major differentiating point against established players like Canon and Nikon. The lens currently retails for ~US$1,500 new, but can be found for ~US$1,000 on used markets.

The optical construction of the lens consists of 11 elements in 9 groups, including 2 ED glass elements, which are supposed to reduce the amount of chromatic aberration. The build quality of the lens is superb and resembles build of other Carl Zeiss designed Sony lenses. The barrel is all metal, as are the focusing ring and the bayonet type lens hood. The lens weighs 995g (2.1lb) and measures 89 x 115mm (3.5 x 4.52in). The lens employs internal focusing mechanism so that the length of the lens remains constant at all times. The focusing ring is decoupled and does not move during focusing, allowing for both auto-focus as well as manual operation. The lens focuses down to  72cm (28.3in), providing 0.25x magnification. The aperture is fully electronic and is controlled from the camera, with the minimum supported aperture of f/22. The lens accepts 77mm screw-in type filters.




Sonnar 135mm f/1.8 ZA is a full frame lens but is fully compatible with all Alpha mount cameras, including all APS-C bodies as well as Sony NEX cameras (when used with an adapter). The EFOV of Sonnar when used on APS-C cameras will be 203mm. The manufacturer's box includes the lens, front/rear caps, metal lens hood, soft leather pouch and registration card. Within the scope of this review, the lens was evaluated on a full frame Sony a850 camera.


Lens Composition 11 elements in 9 groups
Angular Field 18 degrees
Minimum Focus 72cm/2.3ft
Focusing Action AF/MF
f-stop Scale f/1.8-f/22, electronic
Filter Size 77mm
Lens Hood Metal (included)
Weight 995g/2.1lb
Dimensions 89x115mm/3.5x4.52"
Lens Case Soft leather pouch




Sony Carl Zeiss Sonnar 135mm f/1.8 ZA is one of the most interesting lenses I have had a chance to evaluate recently. For starters, Sonnar is the current speed king among 135mm lenses. Both Canon, as well as Nikon offers 135mm f/2 versions in their lineup, but no-one else goes for the f/1.8 aperture. In the days past, particularly when 135mm focal range was more fashionable among users, a couple of other 135mm f/1.8 lenses were available on the market, most notably Pentax smc P-A 135mm f/1.8 in Pentax K mount and Sigma 135mm f/1.8 in T mount (sold under Soligor brand name in the US, and believe it or not also available in MA mount, albeit in very limited quantities). But the absolute speed record in this class actually belongs to Vivitar, which was offering 135mm f/1.5 lens in T mount. But all these lenses are history. The two more appropriate reference points for the ZA Sonnar should probably be the Conatax versions of 135mm lens manufactured by Kyocera up until 2005. The Contax Sonnar 135mm f/2.8 was a fairly inexpensive but a very decent performer, that costs only a fraction of the Sonnar ZA. Contax Planar 135mm f/2 on the other hand cost about what the modern Sonnar ZA costs and was not a particularly popular lens in its days. But, I don't really plan to compare these two lenses, firstly because the ZA lens is clearly a heavily redesigned lens, with more elements/groups, and secondly because adapting a Contax lens to Sony is a pain in the neck - you need to replace the entire mount, which would set you back ~US$100. On top of that, I no longer own Contax  lenses and it's doubtful anyone lending me theirs would be too keen on having their lens tinkered with.

ISO 400, 1/320, f/1.8, 135mm

Ok, enough with the history. Sonnar is one of the bulkiest 135mm lenses I have seen. The bulk (and weight) come from the fast aperture and the optical design, which employs large front elements. The overall weight of the camera/lens combo is rather on the heavy side - while it is still possible to hold the camera/lens in one hand, the front-heavy Sonnar puts quite a bit of pressure on the wrist, so you are likely to find yourself using your second hand to support the lens at the base. The lens does draw attention in public though - expect puzzled and curious looks from passersby when you point that huge front element in their direction. The combo balances decently on a tripod, although I'd recommend using a good ballhead with tight pressure control to prevent the downward drip. On smaller cameras, particularly Sony NEX, the lens looks ridiculous - you basically don't even see the camera behind such a bulky lens. Add a lens hood to that, and well, you get the point...

The lens includes a DOF scale, with markings at f/22 only, making the DOF scale almost useless to anyone planning to use the lens in a preset mode. Considering that the markings are positions fairly close to each other, it is impossible to accurately tell where your focusing range is going to be, probably somewhere in the 10m to the infinity range, but that's the best estimate one can make here. Ultimately, the lens is designed with auto-focusing operation in mind and given that the focusing ring rotates for about 110 degrees going from the closeup to the infinity, the DOF scale is probably an intentional oversight here. And unlike the Distagon 24/2 ZA, reviewed recently, Sonnar does not have an AF/MF switch on the lens barrel, so you would need to switch the camera itself into the manual mode if you decide to focus the lens manually. There is, however, a focus lock button which can also be re-programmed by the camera to activate the DOF preview.

Speaking of the auto-focusing. The AF on the lens is achieved with mechanically driven pin, which locks into the screw slot on the camera mount. This results in a somewhat noisy operation, reminiscent of Nikon pin-driven AF operation, which is also achieved with a mechanically driven pin. The focusing speed is decent, but cannot be compared to the speeds of SSM enabled lenses. Same goes for the focusing accuracy, which particularly suffers at closeup distances, where the lens tends to hund fairly extensively, often locking on everything except what you want it to lock on.



When considering Sony's Sonnar 135mm f/1.8 ZA as my first lens for the then newly purchased Sony a850, I had one 'absolute must have' requirement for the lens - closeup performance at wide apertures. Many considering this lens would probably have similar requirement - the main reason to have a 135mm lens, after all, if the portraiture type work. This requirement basically translates into a good resolution in the center (borders are less of a concern at close distances because of the shallow DOF blurring the OOF areas) as well as good handling of color and contrast. After owning the lens for, ohh like almost a year, I still find myself as excited about using it for any appropriate situation, as I was when I first received it. While I will cover DOF/bokeh and color handling in the followup sections, let's first take a look at how the lens resolves.

Resolution tests were conducted in the 1m to 10m focusing range, giving 7x to 74x focusing distance to focal length ratio. A single set, producing highest MTF results as measured by Imatest was selected and plotted into the chart below. Looking at this simplified MTF50 chart, you can notice that there is literally very little variance across apertures and across the frame. Sonnar ZA produces very even distribution, which is quite rare for fast lenses. And while the absolute raw resolution is not the highest, meaning that it does not shoot off the chart, from visual perspective, it is unlikely users will see any difference across the frame. My guess is that the lens would hold its ground fairly well for sensors up to 30Mp - beyond that we will probably start seeing some drop off in quality at wider apertures.


Sony a850 (24Mp)


The crops shown below compare image quality around borders at f/1.8 and f/8. No matter how long I am eyeballing these crops, I don't see that much of a difference between the two, which sort of confirms the raw MTF output produce by the Imatest - image quality seems fairly consistent across the aperture range, which is what we're looking for in a lens.


Sony a850 (24Mp)

Image borders @ f/1.8 and f/8 (Sony a850)


Before we wrap up with the resolution section of the tests, let's take a quick look at the crops below, taken with a850/Sonnar combo in the field. The lens was focused at around infinity (~20m to the target). The lens was first auto-focused at a distant object, then switched into manual mode to remove any possibility of focus change. I took a series of focus bracketed shots and compared them against each other, looking for the series sharpest at a particular focusing point. Looking at the results, I see minor differences here and there between f/1.8 and f/8. Center looks practically identical save for somewhat higher level of color fringing (more on that later). Both upper right and lower left corners also are fairly close in quality, maybe a little bit less contrasty at f/1.8, but basically image quality at infinity seems pretty consistent and is certainly good enough for most users.













Finally, here's a graph showing the dependency of the image resolution (measured by Imatest) to the focusing distance to the target. Notice the drop off in quality at closer distances due to the softer borders.


Color & Rendering

Having owned a couple of 135mm Contax lenses in the past, I am not 100% sure which one to use as a reference point for comparison against the new ZA copy. As mentioned earlier, technically speaking, Contax Sonnar 135mm f/2.8 is the most appropriate direct predecessor. However, Contax Planar T* 135mm f/2 is probably a more relevant lens to refer to, given its similar large maximum aperture. The Contax Sonnar was decent, but kind of a ho-hum lens, not particularly memorable in any particular area. Planar, on the other hand was somewhat of a controversial lens. It was fairly soft wide open with warm, eye-pleasing color rendering - an good portrait lens, but less universal than the venerable Planar 100mm f/2 or even the Makro-Planar T* 100mm f/2.8. And considering its (surprising) ~50% premium over both 100mm Planars, I chose to sell the Contax Planar 135/2 and invest the $$$ into a lens that would see more usage. Yet after using Sonnar ZA for about 8 months, I can no longer imagine living without this lens, as the ZA variant seems to be superb to the old Contax lineup in pretty much every regard except for size.

ISO 100, 1/2000, f/1.8, 135mm

Without having an opportunity to compare Contax and ZA lenses side by side, and going strictly by memory as well as bunch of sample images left from the Contax days, I'd argue that the new ZA version improved in three major areas - resolution (we already discussed that in the section above), tonality reproduction and out-of-camera color rendition. The last one is probably the most subjective one and given the flexibility of the modern post-processing software is less of an item these days, while the first two are equally important and help differentiate a good lens from an average one.

The contrast levels produced by the ZA Sonnar are probably the most interesting subject here. The lens is not as 'uber-contrasty' as the 100mm Makro-Planar for example, but has a noticeably better reproduction than the Contax Sonnar 135/2.8. At wider apertures, the ZA Sonnar has a nice blend of moderate micro-contrast in areas that are in focus, and reduced contrast in the out of focus areas. Tonality reproduction seems to be slightly better for shadows than for mid-tones, but overall files produced by the a850/Sonnar combo pack enough data to allow you to bring very fine amount of detail in the color gamut (I shoot in the Adobe RGB color space, which provides ~50% better color gamut coverage over the traditional sRGB space, particularly in cyans and greens, but also requires more extensive tweaking in post-processing). At smaller apertures, contrast levels get a noticeable boost and so that the generated images look contrasty from corner to corner.

Chromatic aberration was more or less under control, with lateral CA remaining fairly minimal on the full frame Sony a850. Center CA averaged about 0.2px across the tested apertures, while border CA remained under 0.4px, which is fairly low by all standards. In real life you can still see traces of lateral CA in high-contrast areas, but it is unlikely to cause significant subjective quality deterioration. However, the lens did show some noticeable longitudinal aberration, which, depending on the situation, can become easily noticeable even without the need for high-magnification. The shot of the singer in the park on the right hand side is probably the worst example in this regard, sporting both lateral as well as longitudinal chromatic aberration noticeable to the naked eye.


Sony a850 (24Mp)


DOF & Bokeh

ISO 100, 1/5000, f/1.8, 135mm

Before proceeding with the usual DOF/Bokeh tests for the ZA Sonnar 135mm, I want to make a minor detour and discuss a fairly exotic lens in the Sony's current lineup, which was designed specifically to help photographers achieve the smoothest bokeh possible, without sacrificing image quality. No, I am not talking about another soft focus lens (like the ones available from both Canon as well as Nikon), but rather about the 135mm STF lens Minolta designed in late 90s, which got carried over into Sony's lineup. If the reader is obsessed about bokeh, this lens absolutely must be on the short eval list. I personally have not had a chance to evaluate this lens - it is on my 'TO TEST' list, but the details I was able to pull from the web about this piece of optics are quite intriguing. You see, unlike the typical soft focus lens, which tries to achieve smoother background blur by 'dialing up' the spherical aberration within the lens, the original Minolta STF lens tries to achieve 'better' bokeh by applying an apodisation filter (yea, Minolta certainly did not have good marketing machine at the time), which is effectively a built in concentric gradient filter which decreases transparency towards the edges. The effect is controlled by a separate aperture ring, which opens/closes diaphragm around the filter, hence affecting how the light in the periphery hits the sensor. Bottom line is that the lens tries to give you more creative control over bokeh and loosen the dependency on the aperture level and focusing distance. I am actually looking forward to the day when I can compare the Sonnar ZA to this beast. But in the meantime, let's move on with the regular tests...

One of the more attractive characteristics of the Carl Zeiss Sonnar T* 135mm f/1.8 ZA is its minimum focusing distance of 72cm, which is one of the shortest among 135mm lenses. For a comparison, Canon's 135mm f/2L, an excellent lens on its own, has a MFD of 90cm, while Nikon's aging 135mm f/2D has a MFD of 1.1m. So, effectively, not only Sonnar has a slightly faster aperture among the current 135mm lenses, but it also offers shorter MFD, which in turn should translate into a better separation of the in-focus foreground from the out-of-focus background. 18cm difference might not seem much until you plug the appropriate numbers into one of the numerous online DOF calculators and realize that the18cm difference in MFD increases DOF by 15cm (i.e. the total DOF at 72cm with 135mm f/1.8 lens would be 25cm, while the total DOF at 90cm with the same lens would be 40cm; if we change the f-stop to f/2, to represent Canon's version of the lens, the total DOF increases further to 45cm). Bottom line is that if you're looking for the shallowest depth of field among 'traditional' 135mm lenses, Sony's Sonnar is unrivaled at the moment.

As is customary for us, we're going to start by looking at the shot 'plain, boring' shot of the ruler below, to get a sense for the DOF scale. The lens was pointed at ~45 degrees to the surface, and focused down to ~80cm. You can see how thin the DOF really is, and how quickly OOF areas get blurred as we move away from the point of focus.


ISO400, 1/500, f/1.8, 135mm (Sony a850)


The sample shots below demonstrate DOF in a more or less real-life situation. As indicated in this as well as other reviews, the focusing distance to the subject is where 'the cat is buried' so to speak - you will be able to 'slim' DOF by moving closer to the subject. The two series of shots compare DOF that the lens produces at f/1.8 and f/8 at the MFD of 72cm and then 3m. Notice that when shot at 72cm, the lens achieves almost similar level of background blur at f/8 as with wide open aperture but shot at 3m. Naturally, testing all possible permutations of aperture level and focusing distance is beyond the scope of this review, but these simplified test hopefully give you a hint of what to expect in real life - at f/1.8, the lens would give you a fairly nice, uniformly blurred background as long as you're shooting at short and medium focusing distances. I am speculating that this is where most photographers would be using this lens anyway for head and half-body portraits.


DOF @ 72cm

ISO 100, 1/4000, f/1.8, 135mm

ISO 100, 1/160, f/8, 135mm

DOF @ 3m

ISO 100, 1/4000, f/1.8, 135mm

ISO 100, 1/250, f/8, 135mm


If you have been eyeballing the sample images above at 100% magnification, you might have noticed a minor color fringing in the OOF areas. These are easier to notice in the shot of the ruler above as well as in the shot of the singer in the park, but you can also see some fringing in the shot below with the porcelain cat. But while we cannot technically claim that the lens is resistant to longitudinal CA, the artifact is not very pronounced  and for most practical purposes should not pose problems for users.

Considering that the most typical use for Sonnar as a portrait lens, many users would naturally be interested in the quality of the bokeh rendering. Keeping in mind that bokeh is fairly subjective component, depending on miriad variables, I would nevertheless claim that personally, I have been quite satisfied with the resutls. As hinted above, at short focusing distances, the lens lets photographers to isolate their subjects quite well, rendering backgrounds into an almost shapeless mask - larger objects still retain some basic shape, but are rendered into soft color blurbs with smooth edge transitions, while small objects are completely gone. OOF highlights are pleasantly smooth, with soft edge transitions and even illumination. Contrast levels in the background are fairly subdued, while foreground retains moderate degree of micro-contrast, leaving a nice feel of dimensionality to the objects.


ISO 400, 1/400, f/1.8, 135mm (Sony a850)



To my surprise, Sonnar showed a major weakness against flare. To be fare, all lenses will flare in extreme situations, it's the amount of artifacts that the lens produces in such cases that really counts. The two shots below show how Sonnar performance in one of the more challenging situations, with the sun positioned right above the picture frame on the lens side and the sun rays bouncing off the nearby building before hitting the lens at about 45 degrees. Both shots show fairly low contrast levels, and both showcase noticeable glare as well as color shifts. Nasty...


ISO 100, 1/4000, f/1.8, 135mm

ISO 100, 1/250, f/8, 135mm




The lens registered noticeable, but more or less manageable amount of vignetting on a full frame Sony a850. At f/1.8, the light falloff reaches ~1.1EV, quickly dropping to much lower levels by f/2.8 and from that point, remaining negligible.

From practical perspective, you will see some darkening of corners when shooting light-colored sceneries at widest apertures. Artifact is not particularly distracting and you can easily fix it during postprocessing - set Photoshop's Vignetting Amount in the Lens Correction dialog to +20 to compensate for the falloff. Alternatively, if your camera has built-in vignetting correction, set it to medium level.


Vignetting @ f/1.8 - Sony a850



Sonnar ZA showed negligible amount of pincusion distortion - at ~0.2%, distortion is superficial at worst and will not be visible in real life scenarios.




There are not that many direct alternatives to the Sonnar T* 135mm f/1.8 ZA lens. Sony's own 135mm f/2.8 STF is a specialized lens with defocus capabilities that is somewhat overpriced at $1,200 unless you are specifically looking for this unique capability. Sigma's 105mm f/2.8 EX DG Macro and Sony's 100mm f/2.8 Macro are the only two other choices in the 100mm-135mm range. The only other option is to go for older Minolta lenses, both manual as well as auto-focus. Here, Minolta AF 135mm f/2.8 and Minolta MF 135mm f/2.8 STF could be of potential interest. Another option for those interested in soft-focus capabilities is Minolta AF 100mm f/2.8 Soft. That's pretty much it, unfortunately.



Carl Zeiss Sonnar T* 135mm f/1.8 ZA was my first introduction to the modern Sony system and by far still remains my favorite lens for the Alpha cameras. But without trying to sound sentimental, there's a ton of things to like in this lens. Resolution is absolutely superb throughout the aperture settings, build quality can only be rivaled by other Carl Zeiss lenses (well and maybe Leicas), color rending is excellent, distortion and vignetting are low. Yea, there's an issue with flare, but just use a hood or don't shoot into the sun. The only thing that kind of drives me nuts about this lens is its fairly consistent focus hunting at close distances. Combination of mechanical pin-driven focusing and sub-par focusing detection software in the camera are responsible for the rather iffy results. I have not updated my camera with the latest firmware from Sony as of this writing, but based on the initial reports, focusing accuracy should improve, which would hopefully ease the problem. Bottom line is that Sony Carl Zeiss Sonnar T* 135mm f/1.8 ZA is a strong contender to the crown among 135mm lenses. Now go get a copy before they go up in price...


Sample Images



User Reviews