lens – Ming Thein | Photographer

I’ve actually owned two of these lenses. My first experience was in mid 2010, with the D700 and after discovering the joys of Zeiss microcontrast; I found it stonkingly sharp, very contrasty, yet capable of delivering images with a rich saturation and three-dimensional pop. In other words, very much in line with the rest of the Zeiss ZF lineup.

A study of apples, 1. Inspired by the lighting of the old Dutch Masters. Nikon D800E, Zeiss ZF2 2/100

This lens is my second one – now revisited because I feel the need to find lenses capable of making the most of the D800′s incredible resolving power. Between watches and food, I shoot a lot of macro work. This also means that I’ve got some specific requirements that can only be addressed by a mixture of several lenses; a tilt-shift for increasing depth of field in one plane, or moving the camera out of reflections; something short for use with extension tubes to produce high magnification; something longer to produce better separation/ isolation; and finally, something in a normal focal length that can focus a bit nearer if required.

A study of apples, 2. Nikon D800E, Zeiss ZF2 2/100

I’ve had everything in each category so far until the something longer. The Nikon 105/2.8 VR was my previous choice of all-round macro, but it did have some fairly annoying CA issues that wouldn’t be remedied until stopped down by quite a bit; and the working distance at high magnification was actually pretty short because the non-extending internal focus design necessitated shortening the focal length at nearer distances. I replaced it with the 60/2.8 AFS, which I’ve always felt was a little better optically.

Two-tone pears. Nikon D800E, Zeiss ZF2 2/100

Enter the Zeiss ZF.2 2/100 Makro-Planar T* (hereafter the 100MP). It’s a full stop faster than the Nikon at f2, which it impressively maintains throughout the focus range; it also doesn’t shorten the focal length as it focuses closer, which maintains working distance, as well as minimizes focus breathing (especially important for videographers). The downside, of course, is that a huge amount of extension is required to deliver only 1:2 magnification (extension for a given magnification is proportional to the focal length). 1:1 would have been nice, but I honestly don’t know where they’d pack that extra helicoid. Near focus limit is 44m at 1:2 magnification, with a clear 20+cm of working distance in front of the lens barrel (less if you choose to use the hood).

This is not a small lens when fully extended at maximum magnification with the hood on.

What does amaze me about the 100MP is its ability to cut an image into very clear planes; at every aperture there’s an abrupt transition between in focus and out of focus; in this regard, it reminds me a lot of the Leica 50/1.4 ASPH-M which has a similar ability. This impression is further reinforced by a complete lack of ghosting or fringing of any sort around the focal point, even at maximum aperture. The lens also produces excellent bokeh; out of focus areas are rendered as walls of blur, with no harsh edges or double imaging. The sole exception to this is the occasional cats’-eye-shaped highlight from very bright off-center sources. The iris is placed in the center of the lens’ optical elements, and made up of 9 blades with rounded edges. (The only perfect circle you get is at f2).

Bokeh. Note odd ellipses from off-center. Nikon D700, Zeiss ZF2 2/100

It’s a moderately complex 9/8 optical design, which doesn’t use any aspherical elements (as is traditional for Zeiss) – relying instead on different types of glass and the excellent T* coating to keep optical aberrations at bay.

Flames. Bokeh from the 100MP mostly looks like this. Nikon D700, Zeiss ZF2 2/100

Once again, the coating does its job admirably – flare is very minor, and in fact, almost nil when you use the supplied deep hood; contrast is always excellent, and the microcontrast rendition is superb – very much three dimensional and ‘like a Zeiss’. Color is warm and fully saturated; the lens’ spectral transmission matches that of its siblings, but will probably require some correction if you’re going to use it with those from another manufacturer. And needless to say, as a macro lens, it delivers an almost completely flat plane of focus.

Golf course dawn – shot into the sun. Note complete lack of flare. Nikon D700, Zeiss ZF2 2/100

However, there’s no such thing as a perfect lens – although some manufacturers might claim there is – but the 100MP comes pretty close, in my opinion. Its one sole flaw is longitudinal chromatic aberration caused by uncorrected spherochromatism – in plain text, it’s colored fringes on out of focus highlights (‘bokeh fringing’). It’s especially noticeable front-back on a high contrast subject. The only way to avoid it is by stopping down to f4 or smaller, or some handy Photoshop work with the sponge tool in post processing.

That fruit I can never pronounce. No bokeh fringing because of the relatively low contrast subject. Nikon D700, Zeiss ZF2 2/100

I just want to touch on one last optical property before talking about build quality and some general observations/ conclusions – and that’s diffraction. Although the primary driver of exactly when diffraction kicks in is down to the pixel density of the sensor, I can’t help but notice that there is also definitely some effect caused by the lens used – perhaps this is related to focal length shortening and effective apertures at different magnifications; I’m not absolutely sure. All I know is that if I compare this lens at f22 and the Nikon 60/2.8 G at an indicated f22, the Zeiss does seem a fraction softer – I’d continue the comparison at smaller apertures, but there aren’t any more on the Zeiss.

Sinn 756 S UTC. Nikon D700, Zeiss ZF2 2/100

Moving on to the physical qualities of the lens – like all the ZF/ZE optics, it’s a superb thing to use. The lens is all metal, with a buttery smooth focusing action, and incredibly solid feel. The felt-lined hood locks on to the end of the lens by means of a bayonet mount (in chrome). Let’s just say the lenses feel like instruments, rather than disposable plastic toys. Actually, I do have some criticisms to do with both the cosmetics and the construction, though. Firstly, the red distance markings for feet are too dark and nearly impossible to read unless it’s fairly bright; this holds true for all Zeiss lenses.

Hommage a Monet. Nikon D800E, Zeiss ZF2 2/100

The flocked hood is great at reducing stray light, but it’s also great at picking up lint, and the front edge is easily dentable – if only they would put a small rubber lip on it. I know it’s a macro lens, and the feel is superb, but the focus throw is just much too long – half a turn should be more than enough; the Nikons do this and get to 1:1; there’s no need to have a whole turn of rotation. It makes things slower to use than they have to be. Oh, and despite this huge distance turned…infinity to three meters is probably less than about 10 degrees of rotation.

Mount Yotei Dawn. Nikon D700, Zeiss ZF2 2/100

Finally, it’s a macro lens: for photographing objects, with the lens in close proximity, which may or may not be reflective. The chrome hood bayonet looks magnificent, but it’s also the cause of a huge hotspot (hot ring?) in many shiny objects. The hood helps to some degree, but you can still see the inside of it a little. And that brings me to the nameplate on the lens: white lettering on the black front rim – guess what, this reflects off your subjects too, and has to be retouched out. Again, it’s not as bad when the hood is used (and much, much worse on the 2/50 Makro Planar because of the even shorter working distance of that lens) – but it should be black, or put somewhere else. Better yet, include with the lens a plain matte black blanking ring that covers both the chrome hood bayonet and the nameplate ring when the lens is used specifically for macro work.

Lips. Nikon D700, Zeiss ZF2 2/100

This is perhaps one of the easiest lens reviews I’ve written: the optics are stunningly good, and there are no complaints here. Within it’s optimal working range – it’s fantastic, and longitudinal chromatic aberration aside, can’t be beaten. That said, the LoCA we see here is no worse than any of the other 100/105mm lenses on the market. It’s not only a great macro lens, but it also does very well at longer distances too – I actually like to use it for landscapes, because its tonal rendition really makes scenes pop. You can use it for portraits, but your subjects had better have perfect skin…at least bokeh will be beautiful, though.

Spot the mosquito (hint: click on the image, and look near the ear). Nikon D700, Zeiss ZF2 2/100

If you do any sort of macro work, or are an aficionado of great optics, I can’t recommend this lens enough. It’s one of the few lenses that can keep up with the resolution of the Nikon D800E even at maximum aperture, and versatile enough to serve both as a macro, a portrait lens, and a short telephoto. I’m now off to tape up the front of mine to go shoot some watches. MT

The Zeiss 2/100 Makro-Planar is available here from B&H and Amazon, in Nikon or Canon mounts.

____________

Visit our Teaching Store to up your photographic game – including Photoshop Workflow DVDs and customized Email School of Photography; or go mobile with the Photography Compendium for iPad. You can also get your gear from B&H and Amazon. Prices are the same as normal, however a small portion of your purchase value is referred back to me. Thanks!

Don’t forget to like us on Facebook and join the reader Flickr group!

Spring. Nikon D700, Zeiss ZF2 2/100

Home. Nikon D800E, Zeiss ZF2 2/100

The ZF2 2/100 and D800E produces wonderfully natural color….

…but can also be used as a very painterly tool thanks to its drawing style (D800E, ZF2 2/100).

Waiting for the train in rural Japan. Nikon D700, Zeiss ZF2 2/100

Filed under: Gearhead: Reviews

I realize that in almost all of my past lens reviews, I use terms whose meanings may not be immediately obvious to everybody – allow me firstly to apologize for that, and remedy the situation by explaining some of the more comment terms I use. Note: This is not a technical lexicon (that’s in the works) but rather an attempt to help readers get an understanding of what I mean by ‘luminous transparency’, for instance. I know a lot of you are going to ask for images; I’ll add them in the future once I can find suitable examples. In the meantime, read the text and see if you can spot what I’m seeing in review images, or alternatively, make a small donation via Paypal to contribute towards a straightjacket if you think I’ve gone completely insane. MT

Aberration
Any sort of limitation in the optics that causes distortions in the final image, i.e. straight lines not rendered straight, curved planes of focus, incomplete spectra resulting in inaccurate colors etc – it’s a catchall term for imperfections.

Astigmatism
An effect where the left and right sides of the lens do not render symmetrically, e.g. left edge softer than right edge. Usually the product of misaligned elements. Note that most lenses are capable of theoretically much better results than in reality; the limitation is due to manufacturing tolerances and the impossibility of aligning everything perfectly. Such optical limitations are far more obvious on high resolution sensors, which means that you should really test the specific lens sample you’re going to buy before you actually pay for it…
Note that astigmatism can also be a design flaw where sagittal and tangential rays come into focus at different planes; it looks quite similar to coma. (Thank you Oskar O for pointing this out.)

Bokeh
A general reference to the out of focus areas in an image. Smooth bokeh has no uneven luminance values across the out of focus areas, or hard/ sharp edges, and renders even complex out-of-focus subjects in a uniformly blurred manner. Nervous bokeh is mostly smooth, but can be provoked into double images with the wrong kind of out of focus subject. Bad bokeh is full of oddities like double images, hard edges, artificial highlights, texture in blur areas etc – in other words, it distracts the eye from the primary subject and outweighs it visually. Swirly bokeh is where the background areas are rendered in a spiral pattern as though rotated; this is a trait of many fast lenses – especially the Leica 50/1 Noctilux.

Chromatic aberration
Light is made up of different wavelengths; uncorrected, these focus at different distances away from the lens, resulting in a lateral spread of color across the focal plane. Well corrected lenses use pairs of chromatic doublet elements that bend all of the wavelengths of light by the same amount at every doublet; this ensures that all wavelengths of light focuses in the same plane. Chromatic aberration, or CA, manifests as red-blue fringing on high contrast areas – this is the image separating out into its constituent spectra. Red and blue are usually the most obvious because they have the greatest difference in wavelengths – between 450 and 750nm or thereabouts.

Cinematic
This is one of the tougher properties to describe, and very personal. For me, a lens is cinematic if it has a few properties – smooth background bokeh, and transparency in foreground bokeh; slightly biased but accurate color transmission – usually warm; minimal to zero chromatic aberration; gentle flare; a well-defined plane of focus. These properties usually require at least a short telephoto to achieve, but there are some shorter lenses that do admirably well – the Zeiss ZF 21/2.8, ZF 28/2, Nikon AFS 24/1.4, Leica 35/1.4 ASPH FLE, 50/1.4 ASPH and 50/0.95 ASPH come to mind.

Cold
A lens whose color transmission is biased in favor of the shorter wavelengths, i.e. blues; it attenuates reds.

Coma
The spreading of point light sources; usually at the edge of the frame. Points will not be rendered as points, but instead somewhat egg shaped – a corona of white with a bright ‘yolk’ at the center.

Contrast
The way a lens renders luminance differences in a scene – high contrast means that these are exaggerated; low contrast means that they are minimized. Low contrast lenses are good for high-contrast scenes, because they help to limit the dynamic range of the transmitted image – and thus leave it within the sensor’s capabilities.

Dark
A lens with high saturation and very low flare or internal reflection; ‘darkness’ in a lens mean that colors are rendered in a rich, saturated manner; low flare and internal reflection reduces the overall key of the image and eliminates bright but low-contrast areas.

Distortion
A flaw in the image where straight lines in the subject are not rendered as such. Perspective distortion is when verticals or horizontals that are not perfectly parallel to the imaging plane converge or diverge; lens-induced distortion (barrel or pincushion) is when straight lines are rendered as curving.

Field curvature
An ideal lens projects a flat subject plane onto a flat image plane; most lenses suffer from a mild degree of curvature in both, which is to say that the projected surfaces of subject and image are the opposite sides of a sphere – curving away from the sensor plane and towards the optical center of the lens at the edges. In real terms, this means that the plane of focus in the subject isn’t always the same distance away from the camera.

Flare
A bright spot in an image caused by internal reflections off surfaces within a lens; usually caused by stray non-image forming light coming in at an oblique angle and reflecting off elements; this kind of flare is well-defined and usually lens-shaped. Flare can also take the form of a general lowering of contrast that appears as a white haziness across the image; this is caused by light scattered off the internal surfaces within a lens assembly – including but not limited to edges of individual elements, helicoids, etc.

Flat field
A lens that does not have perceptible field curvature. Most macro lenses and super telephotos are flat field.

Gentle
A rather abstract concept that is used to describe tonal gradations and contrast; shallow output gradation for a given input gradation would be shallow; the opposite would be strong.

Latitudinal chromatic aberration
Colored fringes (multiple images) on a solid edge caused by different wavelengths of light not focusing in the same place.

Longitudinal chromatic aberration
Colored fringing in the foreground and background areas outside the plane of focus, caused by prismatic separation of different wavelengths of light by the lens elements.

Luminosity
Another useful but slightly vague term; my idea of a luminous lens is one which has some minor degree of flare and moderate to low contrast, resulting in a bright or high-key image. Blacks are not always rendered as true black, and images shot with the lens at the correct exposure turn out somewhat brighter than expected. Not necessarily representative of low microcontrast.

Macrocontrast
The projected difference in tonality in the image, relative to the perceived difference between bright and dark areas of substantial actual luminosity differences in a subject. In short: how black are the blacks, and how white are the whites?

Microcontrast
Not quite the opposite of macro contrast – microcontrast is the lens’ ability to maintain small luminance differences in a subject and transfer them to the imaging plane. Note that lenses which have low CA are typically high microcontrast, as there is little ‘averaging out’ of fine detail structures when all of the light from the subject lands up at the same spot.

Neutral
A neutral lens is one without tonal or hue biases in any direction (warm or cold, light or dark). It reproduces exactly what it sees.

Purple fringing
Any portion of the image, usually high contrast, that has a purple glow around it; this is often confused with a lens effect but is actually a sensor effect; it’s due to minor internal reflections off the coatings of the microlens array covering the individual photosites. The color is uniform around the high contrast area, and does not shift left-right as chromatic aberrations would.

Resolution/ resolving power
The ability of a lens to reproduce fine detail structures. Related very intimately to microcontrast – a lens that can resolve high amounts of detail must necessarily be able to resolve small luminosity differences in adjacent areas.

Sharpness
A perceptual effect rather than an actual, measurable property; not to be confused with resolution. Something can be sharp with little resolution – i.e. able to project large differences in macro contrast with little chromatic aberration, creating a sharp edge; but have poor microcontrast resulting in low resolution, but high perceived sharpness. Rather, sharpness is a subjective measure for how much difference we perceive between adjacent areas of different tonal values.

Spherochromatism
The presence of chromatic aberration in bokeh – out of focus highlights separating into distinct red-green portions.

Tonality
The lens’ ability to reproduce luminance and color differences in a scene, and the nature of how it renders these differences in the image.

Transition zone
The plane represented by the distance from the camera where things just start to lose critical focus; a bit in front of and a bit behind the actual focal plane.

Transmission
The lens’ ability to capture light and transfer it to the image plane, limited only by the physical aperture/ iris of the lens. Applies to both the overall amount of light – T stop vs f stop – and the spectral transmission of the lens,i.e. which light spectra the actual types of glass allow through. See also this article on T stops and F stops

Transparency
The ability of a lens not to impart any of its own distinguishing characteristics on the image; it renders the scene as-seen, without any distracting properties or aberrations. My idea of a transparent lens is one with high microcontrast, moderate macrocontrast, natural spectral transmission with good saturation, and a T stop that’s close to its F stop. Bokeh is simply blurred and does not exhibit nervousness or swirliness; foregrounds and backgrounds are equally smooth. Such a beast is a rare piece of optics indeed.

Warm
A lens whose color transmission is biased in favor of the longer wavelengths, i.e. reds; it attenuates blues.

____________

Enter the August 2012 competition: Compact Challenge – here!

If you enjoyed this post, please consider supporting the site via Paypal (mingthein2@gmail.com); Ming Thein’s Email School of Photography – learn exactly what you want to learn, when you want to learn it or our Photoshop workflow DVDs. You can also get your gear from Amazon.com via this referral link. Prices are the same as normal, however a small portion of your purchase value is referred back to me. Thanks!

Don’t forget to like us on Facebook and join the Flickr group!

Filed under: Articles, On Photography

One of the legendary wideangles for SLR users, the Carl Zeiss ZF.2 2.8/21 Distagon T* (referred to simply as the ’21′ after this) is a fairly complex – by Zeiss standards, anyway – telecentric design with 16 elements in 13 groups and a floating rear group for close range correction. As with most of the modern Zeiss lenses, it’s based on a derivative of the older Contax/ Yashica 2.8/21 Distagon (however, that was a 15/13 design). It’s currently available in Canon (ZE, fully electronic) and Nikon (ZF and ZF.2, the latter of which is fully electronic) mounts.

Like all Zeiss lenses, this is a piece of glass that is extremely solid, moderately heavy (~620g) and very well built all-round – it’s like an old-fashioned scientific instrument, in a good way. There is no plastic on this lens; except perhaps some of the baffling at the front and rear. The barrel is black-anodized aluminium, with chrome front bayonet for the hood, and chromed brass rear mount. My two minor complaints about build quality relate to the hood lining and mount – like all Zeiss lenses, the mount seems to wear very quickly, showing brassing after just a dozen or so lens changes. The hood is metal, solid, and locks into place on the front bayonet thread with a reassuring click – and doesn’t move thereafter. However, it also has very crisp edges that are prone to denting, and the felt lining can easily start peeling around the front edge if you get it caught on something. A thin rubber bumper lip around the front edge would solve both problems handily (and if I’m not mistaken, the hoods for some of the Sony Alpha Zeiss lenses have this).

One curious design quirk is the front portion – it reminds me very much of a martini glass. Although the glass itself appears to be able to fit within a similarly-sized housing as the 2/28 Distagon, the front filter thread is a whopping 82mm. I presume this is so one can stack filters without worry for vignetting, but it could also be because some parts such as the hood and bayonet are shared with the similarly martini-like 4/18 Distagon. Unfortuantely, this makes the lens rather cumbersome to pack as it occupies virtually cube-like dimensions.

Boats, stored. Le Sentier, Switzerland. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon

This is, of course, a manual focus lens – apparently the tale goes that all of the autofocus patents are held by the Japanese, except for Hasselblad, who bought technology from Minolta. Needless to say, these patents are being kept very closely guarded; I honestly can’t think of a reason to buy a large portion of Nikon and Canon’s lenses if I could get autofocus with Zeiss. Industry politics aside, the manual focus ring is perfect – spinning freely enough to change focus distance quickly, but not so loose that you can’t set the distance precisely. The amount of damping is perfect. Why Zeiss can’t make all of its other lenses feel like this is beyond me – they’re mostly a bit too heavy in feel for my tastes, especially in a fast moving reportage scenario.

Spiral. Sasana Kijang, Kuala Lumpur. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon

The near focus limit is just 22cm, which makes for some very dramatic closeups indeed; however don’t be expecting fantastic magnification because it is, after all, a 21mm lens – which means you’re looking at 1:5 or so. More importantly, however, is that optical performance across the frame is maintained even at this focusing distance; undoubtedly thanks to the floating rear group that compensates for near aberrations. I personally can’t think why I’d use it at this range, though it might make an interesting lens for food photography on the OM-D – being a 42mm equivalent.

Boutique interior. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon

Sharpness and resolving power are excellent; this is one of the few wides that really does the D800E sensor justice – even into the edges. The center is already extremely sharp from wide open, and the extreme corners catch up at around f5.6 or so. There are very mild traces of lateral CA wide open in the corners with high contrast subjects and the D800E; they’re not visible at all on the D700 and DX bodies. And there’s no odd color smearing, either – everything resolves in the same spatial location, which can’t always be said of wide angles (the Sigma and Voigtlander 20mms come to mind). Interestingly, this lens has an extremely impressive MTF chart* – almost flat by f5.6 – made even more impressive by the fact that Zeiss MTF charts are measured averages not theoretical maximums.

*The interpretation of which will be the subject of a future article.

Dreaming of the high seas. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon

What of the other optical properties? Well, there’s some vignetting; around 1-1.5 stops in the corners at f2.8, but it’s almost entirely gone by f4. There’s distortion, too; up to 2% taking an odd moustache or sombrero-shaped pattern that isn’t so easy to correct manually, but can be taken care of easily by ACR’s built in profile for the lens. Bokeh is neutral and pleasant, with no hard edges; that is, when you can get enough subject separation to see any bokeh in the first place. You’re pretty much going to be at hyperfocal from 2m if you set f5.6.

Gallery. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon

Finally, a quick word on that famous Zeiss microcontrast – it’s present as expected. Microcontrast is the visible result of several optical properties: high resolution; even and high spectral transmission, and as little chromatic aberration as possible. The 21 has all of these things. Resolution and chromatic aberration are functions of the optical design; transmission dependant on the glass types and coatings used. As with all Zeiss lenses, the transmission of this lens is very high thanks to the excellent coatings – note how in the images above, the front few elements mostly disappear; this is due to the surface coatings not allowing reflected light. It’s especially important for maintaining contrast; good coatings manifest themselves in a deep, saturated look that I like to think of as ‘tonal richness’. But I digress: the T stop of the 21 is 2.9, which is just 0.1 stop down from the physical aperture of 2.8. It means that you’re pretty much going to get as much light as you can out of the optical design.

Boats, in use. Lac de Joux, Switzerland. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon

All of these technical qualities are useless if the lens doesn’t produce great images; it does. The 21 somehow manages to be an excellent balance of technical competence and personality; it’s not entirely a transparent lens, but its personality definitely lends a positive influence to any images shot with it. And despite the distortion, you can use it for architectural work uncorrected if you don’t put any straight edges too close to the frame border – in practice, it’s not that noticeable; far more obvious will be whether your camera is level or not. For critical applications, correct the distortion with ACR/ Photoshop. The drawing style of the 21 falls somewhere between the 2/28 Distagon and the 2/50 Makro-Planar; both resolve at very high levels, have excellent microcontrast and transmission, but the difference in ‘personality’ here seems to be related to the amount of field curvature and distortion; the 21 is not perfectly flat field, but it’s pretty close – which is a surprise for an ultrawide.

MRI machine. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon with flash

One of the things I like very much about the way this lens renders – and perhaps more generally about the 21mm field of view – is that it’s about as wide as one can go without the perspective starting to render subjects unnaturally, so long as you carefully place the foreground in your images. It’s wide enough to convey space – especially in tight interior quarters – but not so wide as to appear unnatural, which is a problem I’ve always found with ultrawides. It’s simply impossible to achieve anything approaching a natural-looking perspective with anything wider because of the configuration of our own eyes: 21mm is roughly equivalent to your peripheral vision. Psychologically, our brains just aren’t conditioned to interpreting anything wider on a regular basis.

Solidity and transparency. Sometimes perspective distortion is actually useful to add a bit of abstraction to an image. Sasana Kijang. Nikon D800E, Zeiss ZF.2 2.8/21 Distagon.

I want to add a quick note on compatibility: thanks to it being a telecentric design, the lens actually works very well on smaller formats – Micro Four Thirds, for instance. Carrying say a Nikon FX body, a M4/3 body and the 21 and 100mm lenses gives you an optically excellent and reasonably light landscape kit covering 21, 42, 100 and 200mm – a nice spacing of perspectives. I did this for several early-morning walks in Switzerland whilst on my last assignment, except I had the 1.4/85 Planar instead of the 2/100 Makro-Planar.

One of those falling trees in the forest that nobody ever hears about. Olympus OM-D, Zeiss ZF.2 2.8/21 Distagon

Up until fairly recently, I’d done my wide architectural work with either a Voigtlander 20/3.5, whose low contrast and lack of bite I didn’t like very much; or the Nikon AFS 24/1.4G, which didn’t have the same microcontrast as the Zeiss. All I can say is that I have no idea why I didn’t get the 21 sooner; it’s another one of those truly outstanding lenses that is a must if you’re a wideangle shooter. It isn’t the most discreet lens for documentary work, but the pictorial results are excellent; but it is an absolute no-brainer if you’re an architectural or landscape photographer. Highly recommended! MT

Get your Zeiss 2.8/21 Distagon is available here from B&H and Amazon

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Old factory reflections. Olympus OM-D, Zeiss ZF.2 2.8/21 Distagon

Filed under: Gearhead: Reviews

The Nikon AFS 85/1.8 G (released at the start of this year) is a slightly odd product. Unusually for Nikon, the new version doesn’t cost a lot more ($50, give or take a bit) than the old one; doubly unusually, it isn’t a warmed-over cosmetically-modified version, either. (It’s also now made in China, which might have something to do with it. The old lens was made first in Japan, and then Thailand since 2010.) The 85/1.8 D was a simple double Gauss design with 6 elements in 6 groups; the new one uses a 9/9 optical formula. The elements in front of the iris ostensibly retain the double Gauss formula, but everything afterwards is new. As far as I can tell, the same basic optical principles apply, except every single element in the rear group has been replaced with an achromatic doublet of sorts; not a true achromatic doublet because there’s an air gap between neighbouring elements.

Images from Nikon USA. The D is on the left, the G is on the right. Note far more complex rear group; the pairs of lenses do effectively the same thing as the single lenses in the earlier design. The images are to scale, too – note increase in size. It doesn’t seem to be any heavier, though – and noticeably lighter than the 85/1.4 G.

This complex formula has two benefits: firstly, lower chromatic aberration because there’s that extra element there for correction; secondly, internal focusing is now possible (the previous design focused by moving the entire optical assembly back and forth, like all double-Gauss designs). As far as I can tell, the front and rear elements (possibly more than one) are fixed, and the rest move back and forth. At this point, it’s worth noting that unlike Nikon’s newer optical designs, it’s remarkably free of any exotic technology – whilst the bottom of most lenses now play host to entire essays in abbreviations about Nikon’s lensmaking prowess, the 85/1.8G is remarkably clean. All it has is internal focusing and the silent wave motor – that’s it. There’s no Nano-crystal coating, no ED glass (let alone Super ED glass) and no aspherical elements. Even the new 50/1.8 G employs asphericals!

Untitled. Nikon D800E

Regular readers of my site will know that I was originally a huge fan of the 85/1.4 G, especially on the D700 for it’s sharpness, quality of bokeh and incredible ability to shoot into direct light sources with minimal to no flare. You’ll also know that despite trying multiple samples, I was never quite happy with the performance of this lens on the D800E; mainly due to lateral chromatic aberration wide open, and so-so edge performance. It’s therefore logical to assume that there are optical quality reasons as to why I’m now using the 85/1.8 G instead. You’d be right.

Caution. Nikon D800E

In all of my A-B testing with similar subjects, several samples of both lenses – the 85/1.8 G was consistently sharper in the center at f1.8, let alone f1.4. it’s possible that real resolution was identical, however, the 85/1.4 G showed so much lateral chromatic aberration that it robbed the images of perceptual acuity. It was also sharper at the edges – markedly so, especially on the D800E. It’s worth remembering that at f1.8, the 85/1.8 G is wide open, and the 85/1.4 G is 2/3rds of a stop down. Granted, it’s easier to design a good slower lens than a faster one, but then again, the 85/1.4 G has a huge amount of technology in it – aspherical and ED elements and Nano-crystal coating, for starters. Interestingly, the optical formulae for both lenses are nearly identical.

Apprehension. Nikon D800E

Bottom line: the 85/1.8 G is sharp at every aperture, across the frame, even on the D800E – providing you nail the focus, of course. There is remarkably little falloff in sharpness from center to edge; consistency and microcontrast improve marginally to f4, but it’s already outstanding by f2.8. I only shoot this lens wide open, which should give you some indication of how I feel about the optics. Chromatic aberration under normal situations is almost non-existent; a remarkable performance. Even though this lens has 7 blades instead of the 9 of its predecessor (and 85/1.4 G), bokeh remains pleasing, neutral and smooth. I’ve yet to see any odd artefacts like double imaging or nervousness, but there is a tiny bit of spherochromatism (color fringing) in the out of focus areas. In fact, it’s one of the better-rendering lenses I’ve used in this regard. Color transmission is neutral, per the current crop of Nikon lenses; though the saturation is unsurprisingly not as high as the Nano-crystal equipped optics.

Phonecall. Nikon D800E

You’re probably wondering what the tradeoff is, given the huge gulf in price between the 85/1.4 G and 85/1.8 G; the last line should have given you a clue. It has to do with contrast, saturation, transmission and flare. It seems that the Nano-crystal coating makes an enormous difference to all four; the 85/1.8 G takes a noticeable hit in every area compared to the 85/1.4 G (it still improves on the old lens in every way, however). The problem stems from flare; when you have extraneous light bouncing around inside the lens between elements – a good coating minimizes reverse reflections off air-glass surfaces – everything else suffers. The most obvious manifestation of this is under backlit conditions, of course – especially when there is a bright point light source in the frame. The 85/1.4 G shows almost zero flare; the 85/1.8 G gives an enormously spectacular trail of reflections off what appears to be every single element. This can be pleasingly cinematic for atmosphere or video work, except the lens has no hard infinity stop, which makes focus pulling challenging. For stage/ performance work, it’s a pain in the ass. Unfortunately, the supplied hood makes no difference simply because it can’t block light from entering the front of the lens – and it’s these rays that are causing the problem.

Ugly flare – and this is after correction by burning and desaturation. Nikon D800E

Internal flare also lowers contrast; macrocontrast especially. Microcontrast is a bit worse, but not by much. By f2.8 both lenses are neck and neck here. The knock on effect is a reduction in overall saturation; no surprises here. Perhaps the least obvious, and most surprising side effect is a huge reduction in transmission (read my article on the difference between T stops and f stops for more detail). At any of the wide apertures, the 85/1.8 G transmits between 1/2 and 2/3 stop less light than the 85/1.4 G; this is to say that if both are set to a physical aperture f2, then you’ll find the 85/1.8 G’s required shutter speed for a given exposure to be noticeably lower than the 85/1.4 G. In other words, if you set 1/100s f2 ISO 200 on both lenses, the 85/1.8 G photo will be underexposed by 1/2-2/3 stop. The reason is because a lot of the light entering the lens isn’t making it to the sensor plane, thanks to suboptimal coatings.

That said, it’s still better than the old lens.

Less ugly, more cinematic flare – but still flare. Nikon D800E

The new lens gains a silent wave motor and loses an aperture ring; it’s still plastic, but now the crinkle-finish variety to match the modern bodies and lenses. The plastic type appears a lot less brittle than the 85/1.8 G, though admittedly I’ve never had issues with any of the older lenses other than a propensity to pick up scratches easily. The silent wave motor isn’t any faster than the screwdriver method; it’s about the same, actually – especially on a body with a high voltage built in motor like the D3 or D4. The difference is in precision: it’s a lot easier to move a coreless linear motor in the small increments required to adjust for small changes in focusing distance than a geartrain with associated backlash. In practical terms, you’ll find the new lens a lot more precise than the old one. (It still remains useless if you use an older camera that requires an aperture ring.) The lens also gains environmental gaskets, making it a good choice for pairing with a similarly sealed body.

Solo. Nikon D800E

I thought that it would be easy to write a conclusion to this review; it isn’t quite so straightforward. I’m going to turn it around a bit instead:

Buy the AFS 85/1.8 D if:

You shoot with a manual focus camera, or want to use the lens both on your Nikons and other systems via an adaptor – otherwise you’ll have no aperture control.
This lens is optically inferior to both of the G versions, and not much cheaper than the AFS 85/1.8 G.

Buy the AFS 85/1.8 G if:

Resolution at maximum aperture and CA are important, i.e. you shoot with a D800/ D800E.
You want lower contrast because you shoot with an older, lower dynamic range body
Size and weight are important; the lens is noticeably lighter and a bit smaller than the 85/1.4 G.
Price is important
You like cinematic flare.

Buy the AFS 85/1.4 G if:

You need as much light gathering ability as possible, or shoot frequently under very low light conditions
You shoot into bright point sources a lot
You shoot with a lower resolution body
Notice I haven’t mentioned bokeh yet: the 85/1.4 G is slightly better than the 85/1.8 G, but it doesn’t justify the increase in cost.

The Nikon AFS 85/1.8 G is available here from B&H and Amazon; the AFS 85/1.4 G is available here from B&H and Amazon.

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews

In what appears to be a hideously enormous oversight on my part, I seem to have neglected to review what is ostensibly my most used lens: the Nikon AFS 60mm f2.8 G Micro-Nikkor. As you might expect, I use this lens for the majority of my commercial watch photography. I prefer it over the 85 PCE for images that require high magnification, as this lens natively reaches 1:1 magnification on its own; thus requiring fewer extension tubes to reach even smaller levels of frame coverage.

Nitro Experiment One

Before we start talking about the lens specifically, I would like to debunk some myths about macro vs micro photography: both have to do with the reproduction ratio created by the lens on the imaging medium; it is format independent. Simply, macro refers to 1:1 or greater magnification (i.e. a 20mm wide object in reality would be 20mm or wider when projected on the sensor plane); whereas micro refers to magnification slightly less than this but more than would be encountered during normal photography – ‘close focus’ might perhaps be a more accurate term. Almost nobody seems to get this right online, even the manufacturers; ‘macro’ mode almost never yields 1:1 magnification, and there aren’t that many lenses that achieve this natively. (I suppose Carl Zeiss gets away with it by sounding German and putting a ‘k’ in Makro-Planar – these are 1:2 lenses.)

Water on slate

The 60/2.8 G replaces its predecessor, the 60/2.8 D, both of which are 1:1 lenses; unlike its predecessor, it reaches 1:1 through internal focusing alone, and the lens doesn’t extend – the front element on the G is a lot closer to the front of the barrel, and as a result, offers greater working distance at a given magnification than the D (which has a very heavily recessed front element). The lens has been completely redesigned with a new optical formula; it’s a 12/9 design with aspherical and ED elements, as well as Nikon’s Nano Crystal Coating. It also has a silent wave motor, but no focus distance limiter (oddly, the older version did have this). Focusing is fast and silent, but occasionally the lens does get ‘lost’ – if you’re say at the near focus limit and point it a subject at infinity, then sometimes it can hunt and fail to find focus. A quick tweak of the focusing ring solves this. One thing I have noticed with all of the Nikon SWM macro lenses is that they appear to be very ‘nervous’ when focusing at close distances; they’ll chatter and hunt and rack back and forth slightly. This could be because I’ve got the camera in AF-C most of the time, but it doesn’t really make sense given that everything is static – camera on tripod, inanimate subject. Still, I haven’t noticed any focusing errors, even on the D800E; in fact, this lens is the only one I’ve got that doesn’t require AF fine tune correction on any of my cameras.

Cigar

I also owned the previous version of this lens, and the difference mechanically is night and day; optically, somewhat less so, but the newer version is clearly better. (I suspect part of the reason why the G appears sharper is simply because it can focus more accurately without any of the backlash inherent to screwdriver-focusing lenses.) The biggest difference in optics between the two version are seen in off-center performance – specifically to do with CA – and bokeh. The new lens has very little lateral chromatic aberration; you have to be shooting something very, very contrasty and bright to excite it. For most subjects and shooting conditions, you probably won’t see any lateral CA at all. Longitudinal CA is a different matter – whilst again better than the old lens (and much better than the 105/2.8 VR), longitudinal chromatic aberration is still visible, as are traces of spherochromatism. It’s not a disaster, but it does mean that some work has to be done in postprocessing to remove traces of this – especially on say, white metal watches. On the bokeh front, the new lens has a 9-bladed, perfectly round aperture diaphragm that makes for very smooth out of focus areas; amongst the best I’ve seen, actually – though at normal distances, a 60/2.8 will not yield a huge amount of separation.

Alphabet pasta

It’s worth noting that the lens’ maximum effective aperture at 1:1 is about f4.8; this isn’t because it’s a variable-aperture lens, but rather because additional magnification always results in some light loss. The Nikon lenses and bodies are the only combination that reports this correctly – not that it matters, because the meter takes care of any necessary exposure adjustments anyway. I suppose it might be important if you were to calculate flash exposure with guide numbers, but I can’t think of anybody who still does that.

On the subject of flash, shooting into the light yields no problems at all; the Nano-coated element is clearly doing its job when it comes to suppressing flare. (I use partial backlight quite often to clean out backgrounds or help define the texture in watch dials.) Macro-and micro-contrast are both very good, improving slightly on stopping down. I feel this lens has a bit more microcontrast ‘bite’ than overall global macro-contrast; this isn’t a bad thing at all as it helps to extend dynamic range somewhat.

Breguet La Tradition

I actually don’t have much to say about resolution and optics: what do you expect? It’s a macro lens. There’s almost zero distortion or field curvature, and nothing funny going on with the focal plane. Sharpness is already excellent at f2.8, though with the D800E you’ll probably have to go to f4 or f5.6 to hit peak resolving power across the frame. Note that diffraction softening will set in by around f13 or so with the D800E; I try not to go past f16 unless I absolutely have no choice. That said, you can get away with f22 on the 12MP FX cameras if you need to.

Something I’ve been asked in the past is why I don’t use the 105/2.8 VR instead for greater working distance; the answer is that for the kind of work I do, the 60 actually holds several advantages. Firstly, I don’t need as many extension tubes to achieve higher magnifications*; secondly, the lens itself has much lower chromatic aberration than the 105 – lateral is fairly well controlled on both, but longitudinal is ugly on the 105 – and requires a lot of work to fix afterwards. Finally, there’s the issue of depth of field: for any given aperture, you’ll get more with the shorter focal length**. And given that you’re already challenged to find enough as it is, I’ll take any advantage I can get.

Who could resist a steak like that?

*The more mounts you put between your optics and your camera, the higher the chance of something going out of plane.

**A longer focal length does not mean that you can stop down more before diffraction sets in; that’s a property of the sensor’s pixel pitch, not the lens.

Of course, for those situations when I really need to manipulate depth of field, there’s the 85/2.8 PCE Micro – note it’s a Micro lens, because it only reaches 1:2 – and its full array of movements. That – and an accompanying piece on the Scheimpflug effect and how to properly use a tilt-shift lens – will be the subject of another article.

Girard-Perregaux F1-047

For the work I typically do with macro lenses – watches and food – the pairing of 60/2.8 G and 85/2.8 PCE is usually more than sufficient to deal with any possible scenario. If you shoot bugs, or want the lens to do double-duty for portraits, the 105 is probably a better choice; that’s not to say that the 60 can’t do the job; it just won’t give you the working distance or depth of field control you’d like to have. (The optics remain similarly excellent at longer distances – you could quite happily use this as a long normal lens if you didn’t mind the slowish f2.8 aperture; it out resolves all of the ‘regular’ 50 1.4s and 1.8s I’ve used, especially in the corners.) Perhaps the most telling fact I can leave you with is that of all of the lenses I own, it’s the one that’s been with me the longest. MT

The Nikon AFS 60/2.8 G Micro is available here from B&H and Amazon.

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews

I couldn’t find a product shot in my archive, so you’re going to have to settle for one of me using it instead.

Not long after this lens was initially released and generally available – early 2012 – I published a guest post review here on the Leica Blog. At that point, I’d had no more than a couple of weeks to shoot with the lens, and certainly not under any kind of duress or pressure. Since then, I’ve both encountered many situations with the lens and used it as pretty much the go-to on my M9-P in the hopes of making 35mm one of the intuitive focal lengths in my repertoire. It didn’t stick, and somewhere in the middle of last year, I landed up selling it to one of my students.

Diner. All images in this review shot with the Leica M9-P except where otherwise noted.

I’ve been meaning to do a full review for some time now, but the reality is that there have been many other things which have gotten in the way – or perhaps I should stop making excuses for being lazy.

The 35mm f1.4 Summilux-M ASPH FLE is version seven in a long and distinguished line of lenses – some may even think of them as legendary and quintessentially Leica. They’ve grown larger, heavier and more expensive as time moved on – earlier versions were practically pancakes compared to the 35 FLE, but admittedly they were also relatively poor performers at maximum aperture. The previous version (VI) featured a single aspherical element (there was a very rare double aspherical version produced too, relatively early on in the life of this lens) and was known for being both an excellent optic, but hamstrung by one huge flaw: focus shift.

Saute

Focus shift is when the focal plane changes distance on stopping down, even though the lens elements aren’t moved. This typically happens in fast, non-symmetric lens designs. In practical terms, it isn’t an issue with mirrorless cameras because they’re always focusing at shooting aperture anyway and on the sensor plane; for SLRs, you focus wide open and need to take into account shift when stopped down (DOF preview helps here); for rangefinders, you’re mostly out of luck since you never see where your depth of field plane lies anyway. Basically, the 35 ASPH VI could be calibrated to focus perfectly wide open, or stopped down slightly; at much smaller apertures (8 onwards) it’s a wash anyway as the increase depth of field covers any movement in the focal plane. In my limited experience with the VI, the shift was very noticeable and quite a pain since correct calibration for all of my other lenses would correspond to slightly stopped down for the 35; I landed up selling it and getting the 35/2 ASPH instead, which is both an outstanding optic and exhibits no focus shift.

Reflecting

The new 35 FLE (VII, code 11663) retains the same fundamental 9/5 optical design as the 35 ASPH VI, except for one important difference: a number of elements – the rear group, I believe – move separately from the front elements to correct for focus shift, especially at close distances. Both lenses have concave front and rear elements. As a result of the secondary helicoid, the focusing action is definitely stiffer than the old lens, especially when new. Some use will help the lens to break in and acquire the perfect amount of resistance, however – mine certainly did. (As always, it goes to prove that practice improves your photographic experience…)

Cigar man. From my recent exhibition

The inevitable question is ‘does the FLE group work as intended?’ The short answer is yes – you can stop reading here if that’s all you wanted to know and you were otherwise happy with the older lens. There’s a noticeable improvement in practical image quality close up and/ or wide open; partially because I feel the optics of this version have been somehow refined, and partially because it’s now easy to hit perfect focus even when the lens is used wide open. In several thousand images shot with this lens, I haven’t seen any evidence of focus shift.

Sometimes it’s best not to look up

As far as resolution goes, the lens – my sample at least – exceeds the 35/2 ASPH at every aperture, and seems to have a slightly crisper rendition with improved microcontrast. All in all, an impressive performance. Color rendition is neutral, and takes cinematic color shifts in processing well. Corner performance is almost as good as center performance, with only very slight softness and lateral chromatic aberration visible on high contrast edges when shot wide open; both markedly improve at f2 and match the center at f2.8. Overall, it seems to match the rendition of other Peter-Karbe era lenses very well; I’d say its character is closest to the 50/1.4 ASPH in rendition, splitting scenes into clean, pleasingly cinematic planes regardless of aperture. Subjects stand out with a very three-dimensional rendition thanks to the excellent microcontrast and low presence of CA, though overall contrast seems to be slightly lower than the 50/1.4 ASPH and 35/2 ASPH – not necessarily a bad thing to aid retention of dynamic range on digital. There’s also very little field curvature that I can see.

Sunset over the Gulf of Thailand

The look of modern Leica lenses is defined as much as by the out-of-focus areas as those that are in focus. Whilst a number of aficionados, shooters and pundits alike wax lyrical about the quality of the bokeh defining the various eras in lens design, I think the focal transition zones are much more telling. The rendition of edges in that zone – specifically, the presence of lateral/ longitudinal chromatic aberration and double imaging, and the abruptness of the transition – says a lot about both the design of the lens, and its overall character. In this way, the new-era aspherical Leica lenses all share the same characteristic of having a very fast, clean transition between zones; this is perhaps typified by the 50/1.4 ASPH. I’m sure Erwin Puts and others would have far more detail to add, but this level of understanding is probably sufficient for your average photographer.

Puddle

In practical terms, it means that you probably want to choose a set of lenses based on the look you prefer. Earlier, Mandler-era lenses with shallower, more gentle transitions between zones tend to be appear generally softer and of lower contrast, though this doesn’t mean the in-focus areas have any lower resolution than later ASPH optics. In contrast, the ASPH designs have a much faster, more abrupt – transition, which slices your scene neatly into planes of focus. It’s cleaner, but I personally find suits the digital medium much better than film because the distinct-edges work together with the discrete pixels of the imaging medium to create a doubly-sharp impression. I personally prefer to use ASPH glass on digital cameras and Mandler-era on film.

Occulus. M-Monochrom

I haven’t said anything about the build quality up to this point, because as with all Leica lenses, it’s pretty much a non-issue. The 35 FLE is a heavy, solid lump of machined aluminum (I can only imagine how heavy it would be if they did a chrome version with a brass substrate) that doesn’t clunk or rattle when shaken. The two controls move smoothly; my aperture ring could have used more resistance though; it was a bit too easy to move from setting to setting accidentally. As mentioned earlier, you do need to work the focusing ring to break it in; otherwise using the focusing tab can result in some unpleasant bruises on your finger. I’m actually a huge fan of focusing tabs because they make things so much faster in practice, especially on a well-used and smooth lens.

Fergus Henderson in action

The other huge improvement over the 35 ASPH VI is the hood: no longer is it an enormous clip on plastic monstrosity that both somehow manages to be stubborn (when you want to take it off) and easy to lose/ crack (when you have it in your camera bag) – and to make things worse, it also massively obstructed the viewfinder. Instead, we get a streamlined hood that screws on, managing to stop with perfect alignment thanks to cleverly machined threads; it’s got a tiny cutout in one corner to allow viewing through, but I found that the lens is best used without a hood; simply attach the included blanking ring to cover the hood threads, and use the supplied round cap (instead of that flimsy slip-on plastic thing that covers the hood). In this way, it becomes compact and doesn’t obstruct the finder at all.

Peter Bendheim, good friend and photographer

Out of curiosity, I tried the lens on my OM-D, expecting the same outstanding level of performance as seen on the M9-P; no dice. I was sorely disappointed, and reminded of why I dislike using non-native lenses on other systems, especially those with very short back flange distances. The microlenses covering the sensor form part of the optical system, and tend to interact in strange ways with the optics of legacy lenses. Unfortunately, it was no different here: the 35 FLE showed bad smearing and lateral chromatic aberration (plus purple fringing) and wasn’t acceptable until f5.6 or thereabouts; I can’t recommend using this lens on M4/3 – those who want a fast 70mm will have to look elsewhere.

You can use it for interiors, too!

I’m going to conclude this review by saying quite simply this is perhaps the best 35mm lens I’ve ever used; regular readers and those who know me will also know that I don’t make statements like this lightly. It’s an outstanding performer at all apertures, and there is effectively no penalty for shooting wide open; your rangefinder calibration and eyesight are going to make far more of a difference to achieved resolution than the aperture selected. It doesn’t quite render in the same perfectly neutral, transparent fashion as the 50/2 Summicron APO-ASPH, but it does have the same very pleasing, clean, three-dimensional quality as the newer ASPH glass – I think it would make a great companion to the 50/1.4 ASPH, 75/2 APO-ASPH and 90/2 APO-ASPH, or even one of the wider lenses like the 21/1.4 ASPH (another lens I keep meaning to review) and the 24/1.4 ASPH. If you’re a 35mm aficionado: close your eyes, gird your wallet and buy this lens, then never look at another 35mm again. It’s that good.

Note: half of the images in my exhibition Diametric Opposites were shot with this lens. All of the color ones, in fact.

The Leica 35mm f1.4 ASPH Summilux-M FLE (VII, 11663) is available here from B&H and Amazon

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews, Leica

Nikon’s 80-400mm received a long-deserved update earlier in the year; it’s in fact had a complete overhaul and optical redesign. The original lens was Nikon’s very first VR lens, and body-driven to boot – the large front element had a reputation for pinching fingers between the protruding filter ring flange and the zoom ring (I fell victim to this on my first outing with it). It’s gone from being a 17/11 design to a more complex 20/12, gained Nano-Crystal coating, a shorter minimum focus distance (1.75m in AF and 1.5m in MF vs 2.3m), a silent wave motor and internal focusing, second-generation VR, and plethora of additional switches. Gone is the aperture ring, so you’re not going to be using this on a pre-command dial film body. The hood is also now a petal-type design with the same kind of locking catch as the 17-55, 24-70 and 70-200 hoods. It reverses for storage. Unlike the old lens, it’s also fully gasketed and weather sealed. It’s also more expensive; about $800 more, to be precise.

A grower, and a shower.

The lens surprised me when I unpacked it from what must be Nikon’s largest-ever gold box; it came inside one of those ballistic-nylon zip cases with a shoulder strap, and with a hood. (I’m looking at you, Olympus.) The physical size is larger than you’d expect; it feels quite a bit more bulky than the 70-200/2.8 II, and is a good inch and a half longer than its predecessor, and ~200g heavier. This is not a trivial lens, especially with the hood on and extended to the 400mm position. It’s hand-holdable, but I highly recommend using support of some kind, especially if you’re shooting it with a high resolution body like the D800E. A monopod is fine, and will make a world of difference.

Images in this review were shot at one of my favorite testing and workshop locations in Kuala Lumpur for long lenses – the KL Bird Park. It’s an enclosed free-flight aviary with captive birds, which takes the chance element out of whether you’re going to see any wildlife or not. I used a monopod for support, but did experiment with handholding the lens a few times. Click here for my article on long lens handling techniques.

Too bad the tripod collar is extremely poor, however. It’s flimsy, wobbly, and looks like it was made from two castings melted together, or injection moulded metal; there’s a visible seam line down the middle. It’s easily one of the cheapest-feeling tripod collars I’ve ever seen; unfortunately, it also performs like it – when the lens is mounted on a tripod, the collar doesn’t rotate smoothly at all (it judders) and there’s quite a bit of pitch wobble. This is visible in the form of double images from shutter vibration at marginal shutter speeds (1/125 and below, 400mm). VR helps but doesn’t eliminate it entirely. Fortunately it’s removable entirely, which opens things up to other third-party manufacturers to make a better alternative.

That said, the latest generation of Nikon’s VR system is extremely effective, and is very, very stable once it’s locked in. It also locks in a bit faster than previous iterations and doesn’t seem to ‘jump’ between frames, either. It has automatic panning detection and an active mode that cancels it out. As usual, at high enough shutter speeds – I’d say 1/1000 and above – turn VR off, as you run the risk of double images as the system sometimes has trouble responding at that frequency.

Needless to say, focusing is much, much faster than the old lens; subjectively, I’d say it’s not quite as fast as the 70-200/2.8 II through the full range, but it’s pretty close. (Needless to say, it didn’t even come close to the 400/2.8.) There’s also a two-position focus limiter that allows either the full range or 6m to infinity; this is a rather bizzare choice of distances, as the 6m near limit seems to be not quite close enough for birding, and not far enough to really speed things up if you’re shooting sport. A three position switch would be far more useful – full range, 4m to infinity, and 10m to infinity. There’s also a zoom lock at the bottom to keep the lens locked in the 80mm position, but in practice I found it to be stiff enough not to extend by itself even if carried pointing downwards.

My experience with older lens was from many, many years ago on a D2H and D2x; I do recall that results on the D2x were highly dependent on focusing accuracy, and as with all body-driven lenses, small changes in focus distance were often lost in the backlash inherent in the system. However, it’s been so long that I borrowed something a bit more ambitious for a relative comparison: the AF-S 400/2.8 II. The optics are nearly identical to the current AF-S 400/2.8 G VR II, which is to say, excellent. I’ve shot the latter on several occasions and found it to be one of Nikon’s very best, and one of the best lenses I’ve used, period. It had better be for $9,000.

Whole frame. Difference in rendering between f2.8 and f5.6 at 400mm. Note how the 80-400 also shows vignetting and a slightly different rendering style. I can’t help but wonder if the higher macrocontrast (not microcontrast) is due to the 80-400 having Nano-crystal coating, and the older 400/2.8 lacking it.

Against the 400/2.8, both wide open. Click here for a 100% crop.

The short of the matter is that the 80-400 compared surprisingly well to the 400/2.8; if you compare both wide open, resolution is similar, with the 400 having a touch more microcontrast bite; there’s really not a lot in it, though. Both are critically sharp at maximum aperture on the D800E, with the 400/2.8 improving to peak at f5.6, and the 80-400 improving to f8 and holding constant for a couple of stops as the resolution gains are cancelled out by the onset of diffraction. The 80-400 performed consistently well throughout the range, but being slightly stronger at the 80mm end. One thing I noticed was that the different ends of the zoom required different amounts of AF fine tune – +5 at 80mm, +10 at 400mm (I left it at +8 in the end) to deliver optimal results. No doubt this affected performance slightly, though the difference between a few points of AF fine tune was extremely minimal.

Third leg. Some moments you don’t want to catch…

I found a tiny bit of lateral chromatic aberration, and almost no longitudinal chromatic aberration – a very good performance. If you enable CA removal in ACR, it goes away completely. Bokeh was clean and smooth, with no traces of spherochromatism or hard edges. I did see some very minor double-image effects if there happened to be tight repeating patterns in the scene, but it was never really distracting. Colors are neutral and saturated, with high macrocontrast and moderately high microcontrast; it’s not as good as a prime, but that’s to be expected given the complexity of the optical design. Overall, I’d rate optical performance as excellent – especially so given the lens’ range. It certainly earns a place on the ‘recommended lenses for D800E’ list.

Parrot wing, whole frame – not a crop. This is almost maximum magnification at near minimum focusing distance and 400mm.

As with all of Nikon’s Nano-Crystal coated lenses, flare performance was excellent, maintaining good contrast even when shot into the sun. However, with this many elements, the coatings are only going to go so far in preserving transmission; compared to the 400/2.8 at any given aperture, it seems to lose 1/3-1/2 stop. There’s also some minor vignetting wide open, especially visible at 400mm. Overall though, the optics are surprisingly good: more than up to the task of matching the resolving power of the D800E. I’d be interested to try the lens with one of the 24MP DX bodies to get a better idea of resolving power at even higher pixel densities, but that will have to wait until I have a suitable camera.

In my mind, the 80-400 has several alternatives in roughly the same range: the 70-200/2.8 II with 2x teleconverter; the 70-300/4.5-5.6 VR G; and finally, the 200-400/4. The first option is about the same size and weight (including the converter, smaller without) and gives you another two stops if you don’t need the reach; however, the optics are nowhere near as good as the 80-400. The second option is the lightest and cheapest of the lot, but it has a significantly slower T stop and experiences a noticeable drop in resolving power above 200mm. The final lens is even larger, heavier and more expensive; it’s a ‘proper’ supertele in the same all-magnesium, super-high end build and fast AF mould as the rest of the fast exotics (the 80-400 has a lot of plastics in its external components, presumably to keep weight and cost down); you gain a stop, but boy, does your back pay for it.

One last note on focus breathing: the lens focuses much closer than its predecessor with seemingly not much penalty in size – this is because the lens also shortens its real focal length as it focuses closer, similar to the 70-200/2.8 II vs the original 70-200/2.8. Even at infinity, the true focal length seems closer to about 380-390mm than 400mm (as compared to the 400/2.8, which is 400mm everywhere). At minimum focus distance, I think it’s closer to 330mm; you can see this in the difference in magnification: the old lens gives you a maximum of about 1:4.8 at 2.5m, vs 1:5.7 at 1.7m for the new lens. This might be significant if you don’t want to get too close to whatever it is you’re photographing.

The obvious question is who is this lens for? Indoor sport is out, for two reasons – firstly, the aperture is too slow to freeze action easily unless you’re willing to explore the H-settings on a D3s or D4; secondly, you’re probably going to be frustrated with the focus limits. Outdoor sport – motorsport, for instance – is a pretty good fit; if you’ve got moderately bright sunlight, you’ll get sufficient shutter speeds, and since everything happens further away than 6m, you’ll have extremely snappy focusing, too. What I haven’t mentioned up to this point is that although you should really be using a tripod or at least a monopod for optimal sharpness, you can handhold this lens quite effectively – unlike the 200-400/4 – if you have decent technique and breathing control (at least at 400mm); on a DX body this may be a little different as you’ve effectively got 600mm. Personally, I’d use it on a monopod.

It certainly works for wildlife, as the images illustrating this article demonstrate; however, you’ll either have to use a DX body to get enough reach for birding (the D7100 with its wide-area 51-point coverage would be ideal, I think), DX crop mode, or use teleconverters. The 80-400 is compatible with all of the new TC-E models, though you’re going to lose AF once the working aperture gets smaller than f8. I have no idea what the optical quality of the results is going to be like, as I don’t have any teleconverters handy. I would imagine that it would work quite well with the 1.4x, acceptable with the 1.7x, and don’t bother with the 2x; the 2x takes you down to f11, and by that point you’re already diffraction limited on most of the new high resolution bodies. The naked lens would be just the ticket for a safari, however; decent flexibility, and reach in a pinch. Now, to find the time and money for a trip to Africa…MT

The Nikon AF-S 80-400/4.5-5.6 G ED VR II N is available here from B&H

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews

14-42 X on OM-D, collapsed and extended. If you’re wondering why I got a silver one, it’s because the black ones were out of stock at the time I needed it. Would I have preferred black? Obviously.

I don’t normally review ‘consumer’ grade gear for the simple reason that it’s usually built to a price, rather than built to deliver a certain grade of result (or perhaps it is, only the accountants and engineers know for sure). However, sometimes you come across a piece of equipment that fills a need much better than you imagined; this lens is one such example. The Panasonic Lumix Vario PZ 14-42/3.5-5.6 X G (what a mouthful, hereafter known as the 14-42X) is a very small – about the size of the 20/1.7 pancake when collapsed – zoom for Micro Four Thirds. It was the kit lens for the GX1 and a couple of other cameras for a while, and fortunately also available separately.

This lens is perhaps a surprising choice for me, but allow me to explain the rationale. 1. I needed something compact to pair with a fast prime on the OM-D for what I like to think of as general ‘static’ scenes, where the perspective and framing are of equal importance to the overall success of the photograph; this means that I need to have a zoom of some description. 2. It’s got to be light and small; since I do this kind of photography during the daytime when I travel, and got away with deploying a compact most of the time, it doesn’t have to have a fast aperture. 3. One of the things that frustrated me most about the zoom compacts was inevitably either the lens or the sensor – both have to be of high optical quality, obviously. You never know when you might come across something that makes for good fine art or good stock; no point in limiting the commercial viability of an image because of a quality compromise. 4. A D800E and 24-120 VR would probably be the obvious choice for this kind of thing, but I really don’t want to carry another camera just for this purpose.

I actually rather liked the optics of the collapsing 14-42/3.5-5.6 IIR kit lens that comes with the Pen Minis; however, the build left quite a lot to be desired – plastic mounts and all. Besides, I’d sold mine along with the Pen Mini some time back. The original Panasonic 14-42 has a good reputation, but they’re both large and hard to find; not ideal. At that size, I might as well carry the 12-35/2.8, and that’s significantly more expensive, but not quite fast enough to replace the primes. Enter the 14-42X.

I find the 14-42X to be both a technological tour de force and a bit frustrating. Firstly, it’s tiny – barely larger than the 20/1.7 pancake, but covering a decent 28-85 equivalent (or thereabouts). It shrinks in length by half when closed, and looks positively tiny even on an OM-D. Inside this, Panasonic have managed to pack their usual excellent optical stabilisation system*, Nano coating (more on this later) and very fast AF.

The frustration comes from a different aspect of speed: zooming is slow, and on a lever (though the camera does remember the last used zoom position when power is cycled, a neat touch); and there’s a fraction of a second of added startup delay caused by the lens having to extend. Focus is on a lever, too; I’ve not had any need to use manual focus yet, so I can’t say how effective this is in practice. It’s worth noting that both focus and zooming are completely silent; I believe this was designed with video use in mind. Finally, there’s the odd 37mm filter size and positively teeny (read: easy to lose) lens cap. I don’t even know if any of the good filters are made in this size, then again, I have no intention of using them so it probably doesn’t matter anyway. Presumably one could get closeup filters or something to further increase utility; I’m not sure it’s necessary anyway seeing as the lens focuses to a very useful 0.2m at wide, and slightly more (0.25m or so) at tele.

*It’s unclear to me whether it’s the lens or body IS system in use on Olympus cameras as unlike the larger Panasonic lenses, there’s no switch to disable OIS on the lens. I’m guessing it’s the body system.

You’re probably wondering if the added electrics in the lens have any impact on battery life; yes, they do. But it’s not as bad as you might expect – I’d say for my typical usage cycle, power consumption on my OM-D took about a 20-25% hit; I was seeing full-charge life go from ~900 shots down to 650-700. (I normally turn the camera off after each shot to extend battery life, but in this case it might have actually made things worse as the lens would have to cycle). I do wonder about the longevity of the lens, again given the heavy dependence on (presumably very small) motors to do the work. Then again, it’s cheap and useful enough – around $330 – that if it breaks out of warranty, I’d probably just buy another one.

Optically, the 14-42X is a bit of a surprise: it’s excellent, even used at maximum aperture. You don’t lose any sharpness close up, either. This is important seeing as anything much beyond f8 is severely diffraction limited on M4/3 cameras anyway due to the very small pixel pitch. Use this one wide open without issue, though stop down one stop to 5.6-8 to gain a very small improvement in the corners. There were two aspects of performance I found especially impressive – very low CA, and impressive contrast (though microcontrast could be better). I’m guessing this has something to do with Panasonic’s Nano Coating, and the relatively simple (for a collapsing zoom) 9/8 design. Color rendition is neutral and saturated. There’s no point talking about bokeh, because on a lens like this, unless you’re shooting at minimum distance and wide open, you’re going to get almost no depth of field control anyway.

I think a combination of this lens and the 20/1.7 pancake with one of the smaller M4/3 bodies and would be perfect for the traveller on a strict weight budget, or simply those who want to go fully unencumbered and with just a spare lens in one pocket. These two lenses would let you cover the majority of situations one is likely to encounter, including low light. I’d also recommend it highly as a handy zoom to throw into a bag with a more serious M4/3 kit for the times when you don’t feel like bringing the rest of your gear with you; it turns your camera into a very, very good point and shoot, if you will – no need to spend serious money on one of those large-sensor compacts just yet. Sometimes, flexibility is not a bad thing at all. Highly recommended! MT

More images shot with this lens can be seen here on my flickr stream.

The Panasonic 14-42 X is available in black and silver here from B&H and Amazon.

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews

Announced and available together with the new OM-D E-M1 (reviewed here), the 12-40/2.8 M.Zuiko Digital PRO (24-80mm equivalent) is the first in a new line of M.Zuiko Digital PRO lenses. Development of an equivalent-grade f2.8 fast telephoto zoom was also announced, with a 2014 release. Thanks to the folks at Olympus Malaysia, I’ve had the opportunity to use this lens together with the new camera for some time now. Read on for my review.

Advanced warning: Flickr will apparently be down for maintenance for a little while on Friday 13/9, so if some images don’t appear, it’s because they’re hosted there…

Not a crop.

The 12-40 is billed as being at the top end of the lens lineup. Unlike the previous high grade primes (12/2, 17/1.8, 75/1.8, click on the links for my reviews), the 12-40 is ‘triple proof’ – fully environmentally sealed to match the E-M1, and dust-, splash- and freeze-proof. It has a very nicely made machined aluminium barrel and zoom/ focusing rings, all of which are textured and grippy for use with gloves; overall finishing quality is on par with the primes; unfortunately all of this metal and robustness comes at a price – 382g of weight and bulk (70x84mm, 62mm filter). The lens is larger and heavier than the Panasonic 12-35/2.8, and slightly larger than the 75/1.8 (without hood). The lens also adds a programmable L-FN button for use with your thumb when cradling the combination with your left hand. At the asking price – I’m told in the region of US$1,000 – Olympus have finally included a hood and center-pinch lens cap; both of which are high quality items. The hood is reversible for storage, has a bayonet lock and metal rim; the lens cap appears to be mostly metal and rather nice looking, though I suspect also easily dented.

Accessories aside, the most useful feature that’s made the transition from the primes to this zoom is the manual focus clutch – like the 12 and 17mm lenses, pull backwards on the manual focus ring, and you get both a distance scale and hard stops at either end. This is great for a few things: firstly, taking control of the camera; secondly, pulling focus for video, and finally, if you’re good at estimating distances and depth of field scales, run-and-gun hyperfocal street photography. It’s actually the second item that has me interested. Now that I’m shooting more video, focus pulling becomes an issue; it’s tough with fly-by-wire lenses that lack feel and hard end stops; it’s harder when the speed of the focus pull is oddly proportional to the speed at which you turn the ring, but not the displacement of the ring. The 12-40 (and 12, and 17 lenses) has a neat trick: if you put the ring in the MF position, set your distance, then push it back to AF and focus, it remembers the MF position. This means you can pull focus instantly between any two distances simply by pulling the ring backwards! Better still, the distance is held regardless of the zoom setting. Neat, and very useful in practice. Needless to say, both zoom and focus rings are well damped and have the right amount of resistance for precise setting, but the focus ring is especially commendable.

Autofocus speed is the same as the other recent MSC lenses in the Olympus M4/3 lineup: very, very fast and completely silent. The lens performs pretty well in C-AF mode together with the new PDAF sensor, too. No complaints here at all. What is noteworthy though is that larger physical size of the lens has enabled the designers to include more helicoid; the upshot of which is that the lens focuses down to 20cm from the sensor plane at all focal lengths: in reality, this means about 4cm of working distance from the front element at telephoto, and ~5.5cm at wide. Maximum frame coverage is 48x36mm, meaning slightly better than 1:3 magnification. You’d need a dedicated macro lens on full frame to achieve this. Better still, as we’ll see later, there’s no compromise in optical quality even at this distance.

Optical formula and MTF chart compared to the 12-60/2.8-4 for Four Thirds, courtesy Olympus Malaysia.

The lens has a rather exotic optical formula – 14/9 construction but with one aspherical ED element, two regular aspherical elements, one DSA element, two normal ED elements, one HD element and two HR elements – there’s virtually no ‘normal’ glass in there at all. It also benefits from Olympus ZERO coating (previously seen on the 60/2.8 Macro), whose aim is to reduce flare and increase microcontrast. It works. I did encounter occasional flare in very high contrast situations – visible as a bit of ‘spillage’ around the edges – but the shadowed portions of the frame retained detail, contrast and saturation well. Color rendition was neutral to slightly warm, and richly saturated.

Full test scene.

Corner crop from E-M1 SOOC JPEG, processed through new TruePic VII engine. 100% crops are here.

There’s some minor longitudinal CA on very high contrast subjects, but very little lateral CA in the plane of focus – the worst I saw was about half a pixel at 12mm and f2.8; an excellent performance indeed for any lens, let alone a zoom. You’ll notice in the JPEG sample posted from the E-M1 below that CA is completely absent, thanks to the camera’s new image processor. Regardless of which camera you use, distortion is very, very well controlled indeed – there’s none visible at the long end, and just a tiny hint of pincushion at wide – it actually fares better than many primes in this respect.

12mm test scene.

Center, 100% crops are here.

Corner, 100% crops are here.

Sharpness and microcontrast (related properties) are excellent anywhere in the frame, at all apertures. This is a lens which does not appear to improve much when stopped down; partially because performance is already excellent wide open (it causes the E-M5 to display moire), partially because it seems that we hit diffraction at f5.6, and partially because we have some strange field curvature effects going on. At the 12mm end, center resolution improves by a hair on stopping down, but the edges actually degrade a fraction. The opposite happens at the 40mm end – the center gets a bit softer, but the edges improve. (I repeated this test a few times just to be sure; it could of course be down to my individual sample.) In either case, we’re nitpicking because the difference really isn’t that much; just decide how much depth of field you need and pick your aperture accordingly.

40mm test scene.

Center, 100% crops are here.

Corner, 100% crops are here.

A quick note on close range performance: expectedly, the edges drop in resolution, but the center remains excellent – I was surprised at just how good this lens was as a makeshift macro tool. It obviously doesn’t have the same magnification, microcontrast/ resolution or working distance as the 60/2.8 – but then again it wasn’t optimized for use in this range to begin with. If you don’t plan to do very high magnification work, this may well be all the lens you need.

Note: Given that I was unable to run the E-M1′s raw files through ACR for the time being, optical testing was done with the E-M5 instead.

The lens has a 7-bladed diaphragm with curved edges; it forms a near-perfect circle at most settings, and delivers pleasingly smooth bokeh. The rendering style feels more like Olympus’ primes than what you’d expect of a zoom. And I certainly didn’t see any of the nervous double-imaged backgrounds frequently generated by the Panasonic 12-35, either – and believe me, I was looking for it. (The foliage I was shooting would be the first place this would show up). It separates your image nicely into planes, with a sharp transition between in focus and out of focus elements.

Bokeh and close range performance are both exceptional. This is wide open at 40mm.

In a stroke, I think this lens becomes the defining do-it-all-and-anywhere for M4/3; yes, it’s a bit large, but the useful range, reasonably large aperture, solid build, outstanding optics, very close minimum focusing distance more than outweigh that. It’s not a cheap lens; but then again, I can’t think of any others with the same spec that are. Optically, this is one of the best zoom lenses I’ve ever used. It can replace a couple of primes in your kit quite easily; paired with the 75/1.8, I suspect this will make an outstandingly flexible travel combination. And yes, I’ve ordered one to go with my E-M1. MT

The Olympus OM-D E-M1 is available for preorder here from B&H and Amazon.
The Olympus 12-40/2.8 PRO is available for preorder here from B&H and Amazon.

_____________________________

Enter the 2013 Maybank Photo Awards here – there’s US$35,000 worth of prizes up for grabs, it’s open to all ASEAN residents, and I’m the head judge! Entries close 31 October 2013.

____________

Visit the Teaching Store to up your photographic game – including workshop and Photoshop Workflow videos and the customized Email School of Photography; or go mobile with the Photography Compendium for iPad. You can also get your gear from B&H and Amazon. Prices are the same as normal, however a small portion of your purchase value is referred back to me. Thanks!

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews

This will be the first in my new review format for ‘light’ reviews – pieces of equipment that perhaps don’t necessarily need a full blown magnum opus, but benefit from some context in deployment and typical usage. A short piece on the D4 will follow next.

One of the few lenses in the Canon system I’ve long been jealous of is their 70-200/4 IS (in addition to the 17TSE). Until not so long ago, Nikon users have been missing a light/ compact high quality telephoto option. Sure, there’s been the 70-300/4.5-5.6 VR, but that was only a decent performer up to 200mm; anything else was emergency territory. And it simply wasn’t that good on the D800E, nor a pro build. Finally, we have the AF-S 70-200mm f4 G VR ED IF (what a mouthful). I’m going to address two questions in this review: firstly, is it any good, and secondly, f2.8* or f4? I suspect the latter question is going to be of interest to many still sitting on the fence.

*It’s important to note there are two versions of the 70-200/2.8 G VR. I’ll go into the differences in more detail later.

Inversion

I initially purchased this lens just before the Havana Masterclass; judging from the geography of the place, a short tele would be required to achieve the kinds of compositions I wanted. I already have the Voigtlander 90/3.5 APO-Lanthar, and the Zeiss 2/135 APO-Sonnar; however, neither of these would be flexible enough and frankly I prefer not to use manual focus telephotos on moving objects such as cars and people (both of which there’d be plenty of in Havana). There were three possible options: the 70-200/4, at the same price, a second hand older 70-200/2.8 VR, or the newer 70-200/2.8 VR II. Price is seldom a consideration for photographic tools: if it doesn’t do the job, it’s going to frustrate me to the point that I’ll buy the one that does – especially if I know the option exists.

Armada, 2014

Weight, on the other hand, IS a consideration; so is focus breathing and optical quality. On this basis, the older lens was a no-go. The 2.8 II loses on focus breathing and weight, but has great optics. The fact that it progressively shortens to an effective 135mm at the near focus distance of 1.5m and only hits 200mm at infinity is an issue; you don’t get the magnification you expect and this hinders one’s ability to previsualize compositions. I’m fairly sure that would bother me, so that left the f4. Fortunately, the f4 not only matches – and I think slightly exceeds – the optical quality of its f2.8 sibling, but also appears to have slightly more effective VR, and best of all, focuses closer to a minimum of 1m at all focal lengths, and does not exhibit focus breathing: 200mm at 1m is really 200mm. Why this isn’t the case with the more expensive, larger lens is beyond me.

After the rain

Not everything is a bed of roses though: to achieve the significant weight and price difference – 850g vs 1540g and $1000 – the lens is made of plastic (much like the 80-400 AFS) and there’s no tripod collar or hard pouch included. Plastic construction not necessarily a bad thing (though I personally prefer the haptics of metal); the pouch is moot as I never use them, but making you pay an extra $170 for a mediocre tripod collar is criminal. It’s not that the collar flexes – it doesn’t – but the locking mechanism isn’t very secure; it’s too easy to accidentally release the knob and have the camera rotate. Fortunately, you have to very deliberately pull the knob out and unlatch the collar to release it entirely, preventing expensive noises from ensuing.

Lost

Optically, the 70-200/4 is a 20/14 design, as opposed to 21/16 for its larger f2.8 II sibling. It doesn’t extend when focusing or zooming, it’s weather sealed, equipped with a silent wave motor, and the latest generation of Nikon’s vibration reduction system – supposedly good for up to four stops. In practice, it’s not quite that good, but it does make a huge difference when handholding – especially at the 200mm end, on a D800E. Critically sharp images at the pixel level at 1/125s are consistently possible, as opposed to 1/500+ without. I certainly wouldn’t be without it, and subjectively, it’s probably the most effective of all the VR lenses I’ve used so far. The motor feels like the faster type used in the constant aperture pro zooms rather than the consumer ones;
it feels no slower to focus than the f2.8 II, which is a boon for tracking moving objects. In any case, you should be using this lens in AF-C mode most of the time since small changes in subject or camera distance – especially at the 200mm end – can result in noticeable shifts in the critical focus plane.

A Cuban stereotype

The only thing I have to complain about optically is some fairly pronounced field curvature – the focal plane is simply not flat. It’s not a big deal if you use the nearest focus point, but if you’re focusing and recomposing, you may see some softness and wonder why since most telephotos generally don’t suffer from this. Everything else is good, though: resolving power is excellent even wide open; I’d say every aperture is usable everywhere in the frame, even on the D800E. I think peak is at f5.6-8, but it’s hard to tell as diffraction starts to become visible beyond that. I couldn’t find any signs of chromatic aberration, lateral or longitudinal, which is impressive considering the number of chrome car parts in the sun I shot whilst in Havana. Microcontrast is excellent by Nikon standards – right up there with the best of them – but still falls slightly short of the mighty Zeiss Otus**. Bokeh is always subjective, but in this case, not at all objectionable or busy. There’s a tiny hint of spherochromatism, but it’s not easy to spot most of the time. All in all, the f4 is a lens that splits things up nicely into planes – it has that transparency and separation I look for in my lenses; more importantly, it doesn’t really impose any character of its own onto the scene.

**Is there some bias here? Probably.

Arches off limits

I think we can safely say it’s a good lens; it would appear however that the loss of one group and two elements has done nothing to hurt the optical performance of the 70-200/4 vis-a-vis its larger sibling; if anything, I prefer this optical design for its lack of focus breathing. Close up performance is weaker than distance, especially so at 200mm and f4, but it’s forgivable since it isn’t a macro lens. Most of the time you’ll use it further away anyway, so just set the 3m near limiter and enjoy much faster focus acquisition.

Personally, I’m honestly not sure that the extra stop from the f2.8 II would justify the weight and price difference; I very rarely shoot wide open anyway, and with a lens that long, I’d rather stop down a bit to retain some context; even at f5.6 not everything is in focus most of the time. This is even more apparent with the D800E and an Ultraprint – both are capable of very accurately resolving and reproducing the fine transition between in and out of focus areas, and this should be something that’s used rather than smoothed over with stopping down and aggressive sharpening. Furthermore, I cannot tell the difference optically between the two. The only situation in which I’d recommend the f2.8 is if you are shooting in low available light and/or really need that extra stop.

Untitled

One thing I wouldn’t do again is buy the tripod collar, however. This lens is really meant to be used handheld; it’s light enough and balances very well on the larger bodies – D800E with grip or D4, with most of the weight in the camera – that the tripod collar just gets in the way when you shift your grip. I landed up taking it off. It is a lens which you feel is attached to the camera, not the other way around – I’d be fine with using the camera’s tripod mount if need arose – just remember to turn VR off.

Overall, I’d give this one a solid recommendation for any Nikon users looking for a high quality short telephoto; it’s optically excellent, handles well, focuses quickly, and has very effective stabilisation. Most of the time, I paired it with the Ricoh GR on the wide end and felt that I was missing nothing – in tighter situations, I’d swap it out for the Otus. Or sometimes I’d carry both, with the spare lens on a belt pouch. It proved to be very flexible; it just left me wishing there was a GR with the D800E’s sensor for Ultraprinting…

Neighborhood friends

About the photoessay: This is the first in a series of photoessays from my work shot during the Havana Masterclass in April. I shot a lot of cars, some human context – abstraction of man again – and a bit of cityscape in Havana. Not much of my usual architectural abstraction, simply because the decayed grandeur of the city simply didn’t suit that style. I’m also not much of a telephoto shooter usually; I’d rather keep the context and isolate by composition, or opening up the lens a bit. The geography of the city – long, wide avenues and grid streets – didn’t really suit that way of working because it was too open; you’d either have to get very close to your foreground and use the wide, or compress and layer.

I found myself enjoying the compression/ layering and consciously looking for it; distilling critical elements of the city (cars, people, decaying buildings, arms on windows, Soviet architecture, statues, flags) and attempting to overlay them. Surprisingly, I didn’t do that much cinematic work; I suspect it’s because I’m personally evolving away from that to some extent and moving more towards painterly compositions – and because the dominant colours were so punchy that the impact of my usual tonal shifts would be lost. What you see in light quality and color is pretty close to reality: the latitude, relative lack of pollution and angle of the sun meant that we got the intense color of the tropics, but with the directionality of a higher latitude and strong color shifts between dawn and dusk. Needless to say, all of the images in this review/ photoessay were shot with the 70-200/4 VR on a D800E. Enjoy! MT

The Nikon AF-S 70-200mm f4 VR is available for ~$1,400 here from B&H and Amazon.

__________________

H2 2014 workshops now open for booking – Making Outstanding Images San Francisco, Chicago and Venice; Masterclass San Francisco and Venice – click here to book or for more info

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews, On Photography

One year after the 1.4/55 Otus APO-Distagon, Zeiss is back as promised with the second installment in the new line of super-lenses: the 1.4/85 Otus APO-Planar. Announced unofficially on facebook several months back, the lens makes its official debut at Photokina. I’ve had the opportunity to shoot with a final-pre-production prototype for the last two months; in fact, through pure coincidence, I got the email from my contact at Zeiss saying they had a surprise for me on my birthday…

Gratuitous lens p***.

A huge thank you to the team at Zeiss for the opportunity. I should note in the interests of editorial independence that this review has not been restricted or censored in any way, and there were no usage limitations placed on the lens: in fact, if anything, I was encouraged to push it to the limits. And I did. Test images were made on a Nikon D800E and D810. I’ll be uploading more images to this set on flickr as time goes on. If you recognise some of the sample images in this review from the Zeiss website or stand at Photokina, that’s probably because I shot them…

This is NOT a lens for the weak-armed.

I’m going to call it the 85 Otus from here on, because the full name is a mouthful, and we need to differentiate it from the earlier 1.4/85 Planar that was made in ZF (non-electronic), ZF.2 (electronic), ZE (Canon) and ZK (Pentax) mounts. It’s a derivative of a venerable old design that dates back to the Contax/Yashica days. On the D700 and D3, that lens was one of my favorites because of its extremely cinematic rendering style and flare resistance; it just begged to be shot against the light at every possible opportunity. On the relatively forgiving 12MP sensors, the 1.4/85 Planar was critically sharp in the center even wide open, and not too bad at the edges. Unfortunately, come D800E – we were in for a bit of a rude shock. I actually tried half a dozen samples of this lens before coming to the conclusion that it was simply not usable wide open on the 36MP cameras – softness, lateral CA, longitudinal CA, coma, mild astigmatism – you name it, it didn’t go away til f4. And that very much defeats the point of a fast f1.4 lens. The shooting envelope became very small indeed – and we haven’t even talked about focusing it accurately, yet.

Queenstown at midnight.

Enough resolution for you? This a 100% crop of the above image – remember, its a distant subject with some mild fog, minimal sharpening, and an f2.0 long exposure. Pixel level quality of this combination is not just superb by 35mm/ FF standards, it’s superb by ANY standards – the best of medium format included. It’s quite possibly the best I’ve ever seen, consistently.

I was sad, until the Otus line was announced. And then mildly horrified at the size of the 55 – if that was the normal lens, how big would the short fast ‘portrait’ telephoto be? Turns out the answer is quite enormous. The lens weighs in at well over a kilo, takes 86mm filters, and has a hood that’s about 110mm in diameter. It’s a good 20cm long with hood in place, and honestly, the closest thing in size (but not weight) is the Nikon 200/2 VR without it’s companion hood. This lens is a monster.

Rush hour, Kowloon.

Design and construction are much the same as the Otus 55 (and to a lesser extent, the Touit family) – this is clearly the new design direction for Zeiss. It’s smooth, solid, and for an odd comparison, seriously intimidating in an ergonomic way, much the same as an injection-molded composite firearm such as the FN P90. Everything that appears to be metal is metal. Focusing and aperture rings are made of grippy rubber; my 55 Otus has become not so grippy after six months of heavy use; Zeiss is looking into the issue. There is no plastic on this lens, other than the lens cap.

View from Coronet Peak, Queenstown, New Zealand

Unfortunately, the lens cap design is poor: the springs are too lose, and the plastic is too thin for its size. The two sprung tabs have a habit of coming loose on my lens, and sticking open – upon which the cap just doesn’t stay on. It could use a few more teeth, too. Given the amount of attention paid to every other part of the design, this is a little disappointing (more so, since it’s not exactly easy to find a replacement 86mm cap!). I’ve got one other gripe with the Otus design –it’s that the hole for the focusing scale can collect moisture, which if you’re shooting under adverse weather and very cold conditions can freeze and cause the focusing ring to bind. The water came out again, and I didn’t see any evidence of moisture inside the lens. In all fairness, Zeiss does not claim either Otus to be weather proof – a shame, in my opinion since the strongest images are usually to be made under the most foul conditions. Interestingly, after using the 85 Otus, the 55 feels positively svelte…

Mount Alta before the storm, from lake Wanaka, New Zealand

Peer down the end and you wonder why some of the lens is empty; like the 55 Otus, the coatings on the elements are so good that there’s very little to no reflection, which renders the first few elements pretty much invisible so long as the front is clean. To an aficionado of optics, this is an incredibly sexy lens. Pretty impressive, considering the lens is a 11/9 design – that’s a lot of elements, but to be expected because of the extreme degree of optical correction. This of course means several things: firstly, you can shoot it into the sun without fear of flare (the edges of the elements are coated in the same way as the 55 Otus, too) or loss of contrast*; the T stop is extremely close to the numerical f stop, and color transmission across the entire spectrum is superb. As with the 55 Otus, I’d say the color siguature is moderately saturated but mostly accurate with a very slight cool bias.

*I actually don’t use the hood on this lens because it’s just too fat to pack easily into most of my bags. It’s a good thing it doesn’t need it other than for impact protection, as far as I can tell. I suspect the concave front element has something to do with the lens’ flare resistance, too.

Shotover Gorge, Queenstown, New Zealand

Next up is the easy part: I have tried, but really cannot find any major issue with the optics. There were some instances of what appeared to be magenta/purple edges on very high contrast and overexposed detail, but I don’t know if this is a sensor or lens issue. It isn’t present with proper exposure. Other than that – there’s no visible chromatic aberration, longitudinal or lateral. Contrast is amazing at every aperture, as is microcontrast and ability to resolve high frequency and low contrast detail structures. Resolving power does not appear to improve with with stopping down; it might do, but even at f1.4 it appears to outresolve the D810.

Can anybody identify this galaxy? I searched online for some time and drew a blank. 6 seconds at f1.4 – there’s no CA or coma to be seen anywhere in the frame, even the extreme corners.

I see a slight improvement in microcontrast to peak arounf f2-2.8, but even at f1.4 it’s delivering far more than my Nikon 70-200/4 VR or 85/1.8G does at any aperture. There is some vignetting wide open that disappears by f2.8 or thereabouts, but it’s easily fixable. I did not see any major distortion. All in all: this is seriously impressive performance, and just as good or very slightly better than the 55 Otus. The lens also appears to be fairly flat-field: you must compensate by back focusing slightly if you center focus and recompose for an edge subject.

Between water and rock

On the subject of focusing, I think this is going to be the biggest hurdle for most users: even with my custom focusing screens, calibrated mirror and finder magnifier, it’s extremely challenging to consistently hit critical focus wide open. The viewfinder system is simply inadequate. The only optical finder-based method I’ve found that can nail focus involves racking it slowly while shooting a burst – disastrous for critical timing, but just fine for static subjects. Live view and a tripod is of course highly advisable.

Just before sunset, Arrowtown, New Zealand

100% crop of above.

Most of the time, I used it either on a tripod with live view or stopped down; this is partially because I just don’t shoot wide open that much, and partially because when I do, it’s because I want to use the lens’ ability to separate out distant subjects from the background without any penalty in resolving power. I find this very slight isolation helps immensely with creating a perception of depth in the image. It’s also worth noting that as for the 55 Otus, you won’t have to recompose after focusing: the 85 Otus does not appear to suffer from any visible focus breathing, which will also be of interest to cinematographers.

Tree and mountain – shot wide open at f1.4, the stars are slightly blurred because the plane of focus is the tree. Fortunately a very still night; still enough that the tree blow in the wind at all, and individual leaves are clearly defined.

If the old 1.4/85 was a cinematic lens, the new one is, too: out of focus areas are rendered very smoothly and non-distractingly by the 9-bladed diaphragm, even with high-frequency complex subjects at distances relatively near to the subject. There’s a trace of spherochromatism and bright highlight edges under some very specific situations, but for the most part, it’s invisible. More tellingly, out of focus foreground areas are rendered in a delicate, veiling way: they’re clearly there, and contribute to the feel of the scene, but they never distract. There’s only one small fly in the ointment: for out of focus very distant point sources only, you see concentric ring texture in the big round highlights – this is an artefact of the polishing/grinding process of hybrid aspherical elements (moulded + bonded asphericals). It is only visible under those circumstances, and also present on the Otus 55 – though we seldom notice this because most out of focus sources we photograph aren’t sufficiently point-like. I was told that you either need to have spherical elements that do not require this polishing process, or an extremely costly polishing process (as used on the Master Primes) to eliminate it entirely. Bottom line: not noticeable in 99.999% of normal use situations, but remember, we’re looking for perfection here…

Even horses feel cold sometimes

100% crop of above. Shot wide open at f1.4, very minimal sharpening applied. As usual, if you see haloes…it’s flickr’s downsizing engine.

I’m not necessarily sure the new lens is more atmospheric than the old one, though; I always felt the old lens had a bit of character of its own, but the new one is a very transparent lens – much the same as the 55 Otus. Use them interchangeably at any aperture for a consistent look. Its signature is that it imposes no visual signature of its own (visual signature is usually a result of ‘endearing’ optical aberrations, such as swirly bokeh). This is not a lens whose look you can rely on to add personality to an image: it transmits what is there, nothing more, nothing less. Color transmission and accuracy of rendition is utterly brilliant – slightly better even than the 55 Otus. I was utterly blown away when examining the files from New Zealand on my calibrated monitor. Lloyd Chambers and I are in agreement that this is possibly the most highly corrected lens ever for a DSLR; his detailed evaluation of the 85 is here. There is something in the way the 85 Otus renders that goes beyond even the 55; it is the very definition of ‘clarity’. Subtle nuances are perfectly rendered and you get the feeling of being there. In my mind, this puts it in the very highest class of lenses – the kind that cede full creative control to the photographer, and serve solely to execute their vision. It is an optical achievement of the highest order.

Tree and river

Through this review, you’ll note there are no direct comparisons. I simply don’t see the point. The 85 Otus does things in the corners that no other Nikon lens can do at any aperture; there is no comparison. Even with the Pentax 645Z’s 90/2.8 SR – one of the best medium format lenses I’ve used – there is no comparison. Curiously, the math due to the increased sensor size and extra pixels on the Pentax works out such that the 645Z/90SR and D810/85 Otus combination deliver almost the same pixel dimensions for a subject at a given fixed distance. This means we can almost compare the lens-sensor combination like to like. There’s no question the Pentax/Sony sensor is superior to the Nikon even at the pixel level, especially in dynamic range; however, the 85 Otus leaves the 90 SR trailing. It must be stopped down to f4-5.6 to match the Otus’ resolving power at f1.4 in the centre, and f11 in the corners; and microcontrast/transmission never quite catches up. It’s possible sample variation has something to do with it – I’ve only tested one 90SR, though – they’re rather rare birds here. Still, I think that’s extremely impressive.

Evening.

However, note that it’s not a lens for everybody: dedication to technique and vision are required to extract the most from it. Then there’s the cost and size/weight issue: I’m sure many keystrokes will be wasted to explaining why alternative X at $1000 is better. If you have to even ask why, this is not the lens for you. It may well be comparable if you get a good sample and stop down a bit, or don’t print and only view online; but a big part of the reason why the Otuses (Otii?) are so expensive is because of Zeiss’ QC procedures. I have used half a dozen Otus 55s and two Otus 85s for various reasons and from various batches/ owners – they are all, as far as I can tell, identical in delivering the same extremely high performance. This is not a trivial achievement: any of you who can shoot to the level of maximising everything out of your equipment and have tested more than one sample of a lens will know that consistency is almost impossible to achieve. I have never personally seen any other brand with this level of consistency. If you cannot see the difference (and no web image is going to do it justice; full resolution on a high grade monitor as a minimum, a print ideally), then don’t bother – buy a cheaper AF alternative and not have to deal with manual focus. Extremely shallow depth of field wide open plus that beautifully crisp transition between in and out of focus areas makes achieving critical focus both necessary (missed focus is obvious) and challenging; beyond that, if you don’t have a camera that can make full use of the resolving power and color rendition of the lens, or the skill to deploy all of that potential, it’s somewhat wasted. I honestly feel that the lens still has more to give – but we don’t have the sensors for it yet. I suppose that’s future-proofing. My accountant is already making unhappy noises, but personally, I can’t wait for the next Otus…MT

Coda: I’ve solved the focusing issue with a Zacuto pro finder and live view – stability, magnification, real DOF – just makes the whole thing a bit bulkier, unfortunately.

The Zeiss 1.4/85 Otus APO-Planar is available here to pre order from B&H in Nikon and Canon mounts.

__________________

Limited edition Ultraprints of these images and others are available from mingthein.gallery

__________________

Masterclass Venice (November 2014) now open for booking – click here to book or for more info

____________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews

_Q116_L1030551 copy
The mothership

I was fortunate enough to spend the last three days at Zeiss with Lloyd Chambers (update: his blog entry is here) – with a level of access that I suspect that has never been granted before to independent external parties. They were gracious and first class hosts – I don’t think I’ve had that many types of non-alcohlic beer before. We asked every question we could think of and more, and received answers which we had never expected and at a level of depth that has left me deeply, deeply impressed with what the lens team is doing out in Oberkochen. This may seem like a strange way to talk about the new announcement, but bear with me for while; there is method to the madness. :)

_Q116_L1030556 copy
Origins: it all began with a microscope.

The atmosphere at Zeiss is a very collegial one: there are doors, but nothing seems to prevent both discussion and excitement spreading between all levels of the team. Everybody is passionate and genuinely excited about their product: many are also keen photographers themselves. Whilst we were there, we had access to everybody right up to Dr. Winfried Scherle, head of the camera lens division. Information and ideas are openly shared, and the products are not developed in silos. However, the business of camera lenses is a very small one overall for Zeiss – semiconductors and medical dominate (Dr. Scherle was joking that they are after the decimal point in Zeiss EUR4.2bn annual turnover). It means that there is both pressure to produce products that can match the other divisions in ROI, but also that there is less restriction on developing unusual product that carries a high business risk and potentially small audience (Touits, Otuses).

Zeiss’ Cine lens division is also rightly famous – the lenses have a very high reputation in the industry, and for good reason. They carry a unique rendering that is consistent throughout all of the lenses in each given series (Master Anamorphic, Master Prime, Ultra Prime, Super Speed, CP.2) and in practical terms it means the director and DOP do not have to worry about each scene in a production carrying a different look – this is especially important where there are a lot of fast cuts. Having used the Master Primes myself last year during the Nissan commercial, I can see why. They are assembled at the Oberkochen HQ in an environment that reminds me less of a factory and more of a watchmaking atelier. Lenses are assembled by hand, adjusted by hand, and QC’d rigorously – but with the latest technology. We were lucky enough to tour the inside of the facility, but had to suit up in full clean room procedure before doing so – this is to control dust. First time rejection rate is high and each lens frequently has to go through multiple adjustments (element recentering etc.) before passing through final QC; MTF and a number of other parameters for each lens are measured live on a specially built K8 machine during assembly. Needless to say, tolerances are extremely tight – as one would expect from lenses that can top $30,000+ each.

_Q116_L1030564 copy
Planetarium at the museum – they sell far more of them than you’d expect.

The obvious question now becomes one of relativity: what does this have to do with the price of fish? It is arguable that the Master Prime line is Zeiss’ halo photographic product. Two things filter down through the rest of the lens families: the obsession with quality control, and recently, the visual consistency within a family. It is not just the Cine lenses that pass through adjustment on the K8: Dr. Hubert Nasse, Senior Scientist (and man of infinite knowledge of optics, responsible for the Otus line amongst other things) says there are over a hundred K8s at Zeiss and partners – all used to inspect and adjust lenses on the production line. The difference between say a Master Prime and a Batis or Touit is one of QC: the Master Primes take a day each to assemble and check/adjust; every single Otus is also tested and adjusted with the K8 machine; the rest of the lenses on a reducing frequency. It of course costs time and money to test and hand-adjust every single lens. We were told it is possible to design something with amazingly good optical performance these days with the power of modern computers – but putting it into glass and making several thousand of them at a price which the average photographer can afford is something else entirely. Design compromises have to be made for ease of production; there is no point in having a very ‘sensitive’ optical design that can be perfect but drops to mediocre if one element carries the slightest misalignment.

In real terms, the measured MTF10/20 represent a lens’ ability to reproduce high contrast gross structure of an image; the MTF40 (40 lp/mm) is for fine details – except that was the standard set by the resolving power of film, not today’s digital sensors. The D810, A7RII, 5DSR etc. need 130lp/mm+ lenses. This bring the natural question: why do we not see measured MTF80 or MTF120 values, especially for manufacturers touting higher performance? Dr. Nasse answered this question very simply: such values are so unstable to measure and so critical of machine alignment (each measurement machine has a huge number of moving components that must be micron-perfect to give a meaningful result) and focus that they are not meaningful. The same lens would be almost certain to deliver different results on different machines – but is likely to do so also on the same machine if remounted and refocused. Somewhat worrying when we are effectively attempting this in the field all the time when photographing…and a very good illustration of why resolution gains can fall off very quickly in practice.

_Q116_L1030558 copy
Lloyd Chambers contemplating history

Despite so much being stacked against us, it is still possible to achieve excellent results in practice a high proportion of the time – otherwise I certainly wouldn’t bother with the cost and weight. More importantly, the glass must continue to be able to deliver as time goes on – both from normal wear and tear and as sensors of increasing resolution place higher demand on the optics. I think this is reflected clearly in the longevity and reliability of Zeiss’ lenses: at the risk of being called a fanboy, I admit to having owned more than 30 from the late 50s all the way through today. None of them have gone back for service (except one Otus which had some forensics done on the rubber ring, and a bit of dust removed) – and there have been zero mechanical failures, and only one questionable or inconsistent sample out of all of those. On top of that, many of the older designs still perform well on modern sensors (though not all) and multiple samples are remarkably consistent – this was demonstrated with a series of production Otuses on Dr. Nasse’s own K8 machine – not only was performance consistent between lenses, but consistent across the entire field within the same lens. Lenses are even tested with multiple wavelengths of light (unlike the single wavelengths of a lot of other testing machines) all the way out to the extreme corners. We saw remarkably little skew and almost no astigmatism. That is an astonishingly good performance on all fronts, and the kind of thing that really builds confidence*.

*We also saw the results of lenses from competitors, including some we brought ourselves. The results…clearly demonstrated why an Otus is an Otus.

If you have this degree of a) consistency and b) longevity even within the consumer lens lines – which has been designed in from scratch with the Otus, and the Batis family, to a lesser extent the Classic line, and intentionally with the new Milvus designs – it means that if you can find something that works for you visually, you will be inclined to keep that set of lenses for a long time. Moreover, you will want to use those lenses for a long time. I have found that personally in the Otus line – I just wish they had more focal lengths – and doubt that will change, short of Zeiss making a set of lenses to my specifications**.

**I will have to win multiple lotteries first.

Much has been said about manual focus, and I agree with the current state of DSLR viewfinders, it’s not very practical in the field. Live view helps, but requires a tripod or bulky LCD magnifier. Combine that with two Otuses and something for the wide end, and you need a chiropractor and a Sherpa. Sony has solved this problem for us to some extent with the A7RII. Live view magnification and the stabiliser extend the working envelope of the Otuses – and any MF lenses – enormously; even if the ergonomics are compromised. And it is much easier to place precise focus that way than with AF – there is still a finite size to the smallest AF box, and it can make the difference between hitting eyelashes or pupil. The reality (and I pointed this out in my review) is the shooting envelope of that combination is MUCH larger than anything a DSLR can currently deliver. Note that yes, I am using Nikon lenses on a Sony body. But what if Sony develops its usual product ADD and decides to discontinue the Alpha mount? I would not want to be stuck with E-mount Otus lenses for a dead system; being 100% mechanical and manual ironically adds a degree of future proofing. I know I can use my Otuses on pretty much anything to come – with the same excellent optics, and a consistent look.

_7R2_DSC1252 copy
The first of the Milvus line

I think you can see by this point that the argument for a lens system as opposed to a camera system is starting to make a lot of sense: you have one set of lenses that works on everything, and simply pick the best body for the job. On a tripod, or in a studio setting, I’ll pick the D810 body. In the field, handheld, I’ll pick the Sony. But both will mount Otuses and deliver a consistent look (after profiling the camera, of course). This is massively powerful: but something filmmakers have been enjoying for decades. It is something new to the photographic world, however. Though we also enjoy variety of rendering amongst lenses, I really would prefer to have a consistent rendering that works with my creative vision – if that vision changes, I’ll get new lenses.

_Q116_L1030582 copy
Multiple copies of every lens and different system bodies from Nikon, Canon and Sony were on hand for the media to test

The Milvus is the start of that change at the consumer level: the lenses are also designed to be consistent in color and overall rendering, which is why we have new 1.4/50 and 1.4/85 lenses; the old ones did have a unique rendering but were not consistent even within the rest of the ‘classic’ line (which continues to remain available). Other lenses that do not match will also be redesigned – there is a reason why we now have only six lenses for launch, but far more in the classic line. There have also been significant mechanical changes: aside from a new external design (beware pinching between the end of the barrel and hood, though) and more secure hoods, the lenses now all have full weather sealing, more visible scales, better overall gripability and a selectable clicked or steeples aperture (for video) – there is a small toggle in the rear mount, similar to the Loxia line.

_7R2_DSC0167 copy
Milvus 50 on A7RII

I have been shooting and testing the new 50 and 85 for some time now, and find that the gap to the reference – of course the Otus – has been closed quite significantly. The 50mm is now a Distagon, and whilst it is not apochromatic, performance is significantly better than the double Gauss designs (used by pretty much everybody else, including the earlier Zeiss 1.4/50). Wide open resolution is much higher, peaking at f4-5.6. There remains some secondary longitudinal CA (‘bokeh fringing’) but correcting this would require apochromatic correction and increase cost significantly. It is one third the price of the Otus 55, but achieves 90% of the performance.

_7R2_DSC0201 copy
Milvus 85 on A7RII

The 85mm is even closer: it carries almost the same optical design as the Otus 85, but minus one aspherical element. Again, this results in some secondary LoCA, lower apochromatic performance and slightly lower micro contrast at wide apertures, but an unexpected benefit. The drawing style is somewhat smoother than the very ‘crisp’ Otus 85 and the fully spherical design means very smooth bokeh without any texture in highlights – this is the one limitation of the Otus 85, and actually a reason to own one for portraits and other similar subject even if you have the Otus. I would say it reaches 95% of the Otus’ performance. Unfortunately, physics dictates performance comes with weight: both are still heavy. Both lenses have remarkable flare and contrast performance thanks to the new coatings: they are very flare resistant and maintain great performance even shooting against the sun. Note that for both lenses, Lloyd Chambers already has an excellent and extensive review up – I plan to produce an extended shooting report in the future when time permits, but I admire his patience for testing things I don’t have the time for.

_7R2_DSC0134 copy
Milvus 50 on A7RII. MT is a happy bear right now lens-wise…or at least he will be once the Otus line fills out a bit more…

What we’re seeing is really the start of a significant shift for the serious photographer: the move to mirrorless bringing lens invariance and ease of manual focus is the first part; the second part is family consistency and a sort of focal length invariance for most intended purposes. It means that there is now an even more compelling reason to do some serious research into which of the family ‘looks’ works for your style of photography. The Otuses may be expensive, but if I never have to buy another lens in that focal length again, I think they’re actually a bargain. Choose wisely and that set of lenses can become a lifetime investment. MT

The Zeiss Milvus series is available here from B&H.

__________________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Articles, Gearhead: Reviews

28, 55, 85. A pretty versatile core set for pretty much any purposes. And now fully filled out by the latest in the Otus series, the recently-announced 1.4/28 APO-Distagon. Advance warning: this is not a general purpose lens, nor is it the kind of thing you can deploy casually. That is merely the nature of steeply diminishing returns; there are no gains without significant incremental effort. And we’re really talking about pushing the last 1% here. If you’ve not felt anything lacking in your images, then I suggest you stop reading here and save yourself a lot of money, because chasing perfection isn’t cheap…

Sorry for the gratuitous p***, but it really deserves a place amongst the legends.

Disclaimer: I have been testing the 28 Otus for the last couple of months courtesy of the folks at Zeiss. The sample I have is not 100% final, and I’m told the units that ship will carry further performance improvements. I plan to update this review accordingly when I have access to those lenses. In the meantime, I’ll be uploading images (including 100% screen shots from various parts of the frame) to this flickr album as I shoot with it more. The full data sheet including block diagram and MTF is available here from Zeiss. Additionally, Lloyd Chambers also has an ongoing detailed technical analysis which comes to much the same conclusions as I do.

Extreme corner crop. Note highlight handling and complete lack of chromatic aberration. It’s also worth noting that whilst not all subjects have high frequency detail that shows off the resolving power of the lens, you still need a high degree of transparency to accurately render the small tonal transitions in a low contrast subject like mist or cloud and avoid it looking flat and ‘dead’.

Regular readers will know that I have reviewed the 1.4/55 and 1.4/85 Otii in the past, and have been impressed to the point they now serve as reference lenses for anything approaching that focal length, regardless of subject distance (!). They match or better my macro lenses when used with extension tubes. However, you’ll also know I never carry more than one of these things at a time: it’s just not practical in the field unless you are in a studio environment, are shooting somewhere with direct car access, or you can use a rolling bag.

I also make no secret of the fact that I’m a very big fan of the 28mm focal length – wide enough for context, long enough to still be plausibly natural in rendering. There have been a few lenses that satisfy for various reasons – but optical perfection has never been one of them. Zeiss’ own 2/28 Distagon has been a favourite because of the 3-D rendering caused by serious field curvature; however it suffers from noticeable violet fringing wide open and both longitudinal and lateral chromatic aberration on high resolution cameras. It is fine from f5.6 onwards, but that limits applications quite noticeably. The field curvature and violet fringing make it difficult to focus accurately, too. The Leica Q has a solidly-performing lens that’s great in the centre, but weaker in the corners til f5.6 or thereabouts and relies on fairly heavy software correction to moderate distortion; on top of that, there’s some focus shift. The Ricoh GR and GRII is good out to the corners, but tops out at f2.8 and is fixed to one camera body (with attendant limitations).

Whilst all of these lenses excel at a specific purpose, I find I’m often left wishing I’d also had one of the others in the bag when other opportunities arise. I haven’t mentioned other methods of getting to 28mm-e simply because none of them are particularly noteworthy – I’m sure there are probably some lenses I haven’t tested (like the 28/1.4 Summilux-M ASPH) which might fit the bill, but these aren’t exactly practical, either. Until this point, there hasn’t really been a truly reference-grade versatile 28mm offering that can really be used for anything.

Enter the 28 Otus. Because in reality, nothing is perfect, I’m going to start with the warts first: some are obvious, some are not.

At 1.35kg, a 16/13 design and a 95mm front filter, the spec sheet reads more like a 70-200/2.8. (In practice, it takes up about as much space in the bag, too.) That is a lot of glass, which aside from weight means potential QC and alignment issues – especially for corner performance at large apertures. My sample is remarkably symmetric (as is the other sample I am aware of in the wild), so perhaps this is not likely to be an issue in practice – the level of QC achieved with the other Otii has been remarkable thus far.

This is not a lens you carry trivially: you need to know you a) need 28mm, and b) need the performance envelope that other lenses cannot deliver. It is not a lens that is quick to shoot, because precision in focusing is required (more on this later). It is not a lens that is easy to carry, nor does it balance well on small bodies – the D810 really benefits from having the vertical grip attached, taking the full weight of your rig into the 3kg range, and that’s before you add an LCD magnifier for precise focus, too. It does not balance well on mirrorless bodies – don’t even think about using this on the A7RII without the vertical grip, because your fourth and fifth fingers will cramp very quickly as they try to counteract the torque from the lens on the small grip. And thanks to the adaptor, all of the weight is very far forward. But at least you’ll get stabilisation and a true DOF representation.

Focusing on the fly is not easy. It’s not as bad as I expected it to be with a wide angle lens (read: shallow DOF transition between in and out of focus) and the D810’s optical finder (albeit modified with split screen and realigned mirror) – but it’s not easy, either. And you still need to do a bit of focus bracketing to get it right. Naturally, live view with an LCD magnifier is much easier, but slower since you have to go to 100% at your point of focus to ensure critical sharpness. It’s not quite so simple, however. The 28 Otus has a relatively flat plane of focus with some curvature towards the edges; it’s flat enough that center-focus-and-recompose will not work, and fast enough at f1.4 that you will not have enough depth of field to cover the effective shift in subject distance, either. This means you must focus at the subject’s final position in your composition. On top of that, the nature of the field curvature changes slightly as you stop down – ‘uncurling’ is perhaps the best description – and the plane of focus itself seems to have a larger sweet spot in the centre of the image than towards the edges. For work where you need critical sharpness from corner to corner across a range of distances, I’d focus at the near corners first and then tweak for the centre afterwards. If your scene topography isn’t concave towards you, then focus as normal.

If this seems like a lot of hassle, it’s only because misses are noticeable: that is the nature of high resolving and highly chromatically-corrected lenses. What you get in return, however, is quite spectacular: impressive resolving power in the centre of the frame even wide open; equally impressive resolving power in the corners once you get above approximately 5m subject distance, and performance peaking around f4-5.6 across the field. By f8 you are assured of sufficient depth of field for a wide range of subject distances and effectively flawless performance; the corners are especially impressive. Subjectively, you’re already getting 85-90% of the maximum performance at f1.4. The lens reaches a full 10/10 across the entire frame even on the 42MP A7RII by f5.6. Note that whilst the centre remains strong to minimum focusing distance of 0.3m, the extreme corners are good by ~2m don’t really come on song til >5m as noted before.

Close range, f2, center crop

Resolving power of course isn’t everything – there are other optical characteristics that define the performance of a lens, some much more difficult to define but yet highly visible. Vignetting is of course present given the lens’ speed; this is a mechanical limitation on the size of the front element. There is around 1.5 stops in the corners and this is of course easily corrected. I notice some trace secondary longitudinal chromatic aberration on very high contrast edges, though much less than with any other wide lens I’ve seen. Lateral chromatic aberration is almost entirely absent. Distortion is also impressively low – a little over 1% – given the speed of the lens, and follows a simple curvature that is easy to compensate for in post.

Center crop, f1.4

Whilst I find the 28 Otus does not quite have the same ability to ‘slice a subject into planes’ as its longer 55 and 85mm siblings, I attribute that more to the nature of the focal length than the optics. Again like it siblings, it does have the nature towards somewhat nervous bokeh if you have a high frequency subject that’s very close to the focal plane – this is a consequence of the aspherical elements and an unavoidable tradeoff when attempting to maximise both speed and performance. If there’s one weak point in the 28 Otus’ performance, it’s with wide apertures, strongly backlit subjects and not much subject-to-background separation – this can exacerbate both visible fringing and nervous bokeh. In all fairness though, I cannot think of any wide lens that excels under these circumstances.

All of the Otus series lenses employ an elaborate layering of coatings and baffles to minimise reflections, maximise transmission – and thus both micro- and macro- contrast – and there is no question that Zeiss have been successful here. The 28 has the same ‘bite’ as its 55 and 85 siblings, and there is a very strong family rendering. It’s difficult to describe exactly, but I’d classify it as highly transparent, tending to slightly cool, with good saturation and a clean/sharp rendition rather than a smooth one. If this fits your style – it does for mine – you’ll probably never need to buy another lens in this focal length again. Assuming your bank and back can take it, of course.

Construction quality is first class, as with the other Otii; all mechanical movements are internal to the lens, which does not change external dimension when focused (nor does the front element rotate). There is no environmental sealing, but I’ve used all three of them under unpleasant conditions and not found any issues to date. Looking at a half-cut lens, there’s so much baffling and sun complex helicoid arrangements that I suspect water is going to have a tough time penetrating anyway. It’s also interesting to note that in places (specifically, towards the front of the barrel as it flares outwards to accommodate that 95mm filter – necessary to avoid vignetting) is some air space between the helicoid and the external wall; I suppose this will help for both impact protection and thermal expansion (!). Don’t laugh, if you shoot with one of these things in tropical sun for a few hours, expect it to get very warm indeed. Lastly, the focusing ring’s smoothness is revealed: the part rides on several teflon roller bearings, similar to the Master Prime line of lenses. That aforementioned focusing ring can be a little too featureless/smooth at times – I found adding a small indexing tab to it helped enormously in being able to instantly gauge the focused distance, as well as for something to grab. It’s also worth noting that the 28 – again, similar to the 55 and 85 – exhibits very little focus breathing; something which videographers will undoubtedly appreciate.

Extreme corner crop – it’s sufficiently well corrected to cause moire in the extreme corners of a relatively low contrast subject. Hats off.

Given the non-final state of engineering, I believe it is perhaps a little premature to form a decisive conclusion – however, it is clear that already the 28 Otus has no competitors – and at the expected price and size, it’d better not. The bit that troubles me is deployment: this is a lens that takes up two ‘normal’ slots in a bag (the 85 can still be shoved into one) and cannot be casually deployed with the expectation of reaching maximum potential. This is a deliberate lens, and I think the images I’ve shown here reflect that. That means attendant tripod, possibly LCD magnifier and DSLR body, further elevating weight. You can of course use it on an A7RII, the lens really needs a tripod mount (even for a D810) and the weight of the whole combination is putting a risky amount of torque on the camera body’s mount. I suppose this is also true of all of the other Otii: though the theoretical envelope of the lenses is about as wide as it gets – any aperture, pretty much any distance. Do we have a new reference king? No question. But like a formula one car, making the most of is going to require a little planning and skill. MT

Most of these images were shot during the Chicago Masterclass last month – I have one final place available for the Tokyo Masterclass in three weeks; booking and information here. Images in this review were processed using Photoshop Workflow II.

The Zeiss Otus 1.4/28 can be pre-ordered here from B&H.

__________________

Ultraprints from this series are available on request here

__________________

Don’t forget to like us on Facebook and join the reader Flickr group!

Filed under: Gearhead: Reviews

_8B36984 copy

Following a couple of recent email exchanges I’ve had, I thought I’d tidy things up and publish them here for the benefit of the general readership. This is a list of what I consider to be the ultimate lenses, on their native systems (and irrespective of system, actually). Lenses also tend to have significantly greater longevity (especially if without electronics) especially compared to camera bodies; you could buy one set of Otuses and adapt it to just about everything now and to come. In that sense, whilst good glass is expensive – the long term cost of ownership is significantly less than cutting edge bodies, and given residuals are high, generally worth the investment.

Of course, what constitutes ‘ultimate’ is actually highly subjective; some value smooth drawing quality and tonal transitions over outright resolution; others require zero distortion or high color accuracy or secondary color correction. If anything, my personal preferences tend to lean towards resolution and microcontrast; I can accept some vignetting, distortion, secondary lateral CA (but not longitudinal) – because these are easy to fix in post. Field curvature, smearing, coma etc. are not. Not all lenses on this list are here because of technical perfection or MTF chart performance, either. On top of that, there are two lenses that are not system options, but included anyway because they deserve honourable mentions. There are probably also better lenses I’ve not used yet (and so obviously can’t include them). I’ve tried to give justifications where possible. With that in mind, and in no particular order, here we go.

**Items denoted with two stars are lenses I currently own. *One star, lenses I’ve owned. Links are to reviews or affiliate suppliers. Images shot with the respective lenses mentioned.

Otus 85, D810

The Zeiss Otus series** – in particular, the 1.4/85 APO Planar (review B&H) and 1.4/55 APO Distagon (review B&H Amazon).
I can’t think of anything else that reaches this level of performance across such a wide envelope, from the extremely challenging zone of maximum aperture all the way up to where diffraction kicks in. The 1.4/28 APO Distagon (review B&H) is still the best wide angle there is, but it doesn’t quite reach the levels of its brother. To get an idea of what is possible when budget and size considerations are secondary, the Otuses are it. At the time of writing, they’re the only lenses that manage to out resolve every camera they’re mounted on – by a large amount.

Zeiss 2/135 APO, F2 Titan, Delta 400 +2 push

The Zeiss 2/135 APO-Sonnar (B&H Amazon)
Another best in class: very difficult to focus because of the relatively short throw and abrupt transition, but oh boy – what a rendering! It manages to do this without any aspherical elements, which makes for very smooth out of focus areas and transitions – despite that apochromatic ‘bite’. For the bokeh fans, and those who need isolation at distance.

Zeiss 2/28, D700

The Zeiss 2/28 Distagon* (review B&H Amazon)
Despite having used a huge number of 28mm lenses, my favourites remain this one (for rendering, color and three-dimensionality and separation due to field curvature); the Ricoh GR’s lens (for outright resolution) and the Leica Q’s for overall balance between the two. It’s not expensive, it’s small, but it’s tricky to focus at the edges because of aforementioned field curvature, and it doesn’t play nice on mirrorless because of edge ray angles. But the results are gorgeous…

Voigtlander 180 APO Lanthar, A7RII

The Voigtlander 180/4 APO-Lanthar** (review)
When people ask ‘why small and slow?’ – this is the lens I cite. Barely 10cm long and with a 52mm front filter, it’s almost impossible to believe the focal length. Color correction is phenomenal, and whilst overall contrast is somewhat low compared to modern designs – it does make for very smooth tonal transitions. Micro contrast is excellent, but watch out for flare. Use of the hood is a must.

Nikon 24 PCE, D810

The Nikon PC-E 24/3.5** (review B&H Amazon)
I think this is actually a very misunderstood (and consequently, misused) lens. The problem is that it suffers from a) field curvature b) focus shift c) a very, very sensitive focusing ring, especially near infinity, and d) a movement it doesn’t need, but can’t be zeroed out consistently. The trick is to use it wide open and focus with live view, only caring about your focus point; or at f11, and focus at the most distant thing you want in focus – the shift means everything in front will be sharp, too. Field curvature means the focal plane is like a dome over the lens, which is actually good for most urban situations. You don’t really need tilt on something this wide – by f11 everything from 1m away is critically sharp, even on a D810; all it does is not lock properly at zero, frequently resulting in softness due to unintended movement.

Nikon 85/1.8G, D4

The Nikon AFS 85/1.8 G** (review B&H Amazon)
An odd choice considering the others on this list; but I actually like the way it flares (extremely so). It’s useful for atmospheric images, cinematic feel, and isn’t harsh or distracting. If you ensure there are no point or bright sources in the frame, contrast and resolution are excellent. Note: it has not aspherical elements, which might have something to do with it. It’s also cheap, light, and small. You can put and not notice it’s there til you need it.

Nikon 70-200/4VR, D800E

The Nikon AFS 70-200/4 VR* (review B&H Amazon)
Another one on the list because of size, weight, versatility and pure optical competence – at f4, it’s easy to make a truly outstanding lens, and it shows. I think it actually handles better than the 24-70, in my opinion. And the VR is excellent.

Nikon 24-120VR, D800E

The Nikon AFS 24-120/4 VR** (review B&H Amazon)
With caveats: if you get a good copy, it should be very good everywhere at f4, and quite excellent by f8 post-correction (CA, distortion, vignetting) – even on a D810. It isn’t outstanding anywhere, but it doesn’t have any major flaws, either. Perhaps the ultimate Swiss Army Knife choice; ironically, pared with one of the D4 or D5 series bodies, there really isn’t anything you can’t shoot (and they’re also less demanding than the D810). I’ve shot far more with this than I care to admit…

Nikon 45P, D750

The Nikon AI 45/2.8 P**
There’s something about the Tessar formula that renders rather nicely for portraits; it also helps that its small size doesn’t intimidate, and it’s easy to focus. Another one of those lenses that would make my small, lightweight kit – stop it down to f8 and it’s great; wide open it’s smooth. A bit of a chameleon; sadly no longer made but reasonably easy to find on the secondary market.

Canon 40STM, 5DSR

The Canon EF 40/2.8 STM* (B&H Amazon)
Small, sharp, great bang for the buck – with a rendering style similar to the Nikon AI-P 45/2.8, but with autofocus. Think of it as a body cap that you can also use to take some very nice pictures. What’s not to like?

Canon 70-300L, 5DSR

The Canon 70-300/4.5-5.6 L IS USM* (B&H Amazon)
For the kind of work I do, I preferred the added reach/versatility of the 70-300L over the speed of the 70-200; it’s just as usable handheld thanks to the great stabiliser and relatively small size; it’s also much easier to pack. The tripod collar is a bit weak, though, and you need to watch for vibration. If only Nikon or Hasselblad made something like this…

45/1.8, E-M5

The Olympus 45/1.8 for M4/3* (review B&H Amazon)
Another one of those small, light bargains. I find it needs a little stopping down to have truly satisfying bite, but its character remains smooth throughout the aperture range rather than crisp. It has a very nice rendering style and would be great for video if it had mechanical manual focus, but at the price, I think we probably can’t complain.

75/1.8, E-M5

The Olympus 75/1.8 for M4/3* (review B&H Amazon)
One of the best short teles I’ve use for any format. Performance wide open is already outstanding, including secondary color. There’s almost no longitudinal CA, which is surprising given the speed of the lens and my general avoidance of using a hood. Again, a shame you don’t get the proper AF/MF focus clutch at this price; it has a very nice rending for video, but pulling focus is nearly impossible…

CY 100-300, 5DSR

The Contax Yashica Zeiss 4-5.6/100-300 Vario-Sonnar**
This zoom has a reputation for being better than the Zeiss’ own primes in that focal length and range – and it is. It’s tricky to use because it’s a push-pull design, but very, very smooth and a joy to pull focus on. An aftermarket tripod collar is a must – there is simply no way to get a sharp/stable image out of it otherwise. Overall contrast is low – like non-aspherical designs of that era – but relatively good micro contrast means you’ve still got a lot of information to work with. As with all tele zooms, performs better at the wide end.

85/2.8 MMG, A7RII

The Contax Yashica Zeiss 2.8/85 Sonnar MMG**
Yet another one of the chameleons on this list: lower in contrast and bite wide open, rivalling the Otus 85 in resolving power and microcontrast when stopped down. I converted my version to Nikon F; it’s small size and fantastic performance have made this my choice of 85mm unless I need one of the specific special attributes of one of the other lenses. No aspherical elements in this design either, which makes for very smooth transitions and bokeh.

CY 35 PC Distagon, 5DSR

The Contax Yashica Zeiss 2.8/35 PC Distagon AEG**
An outright impressive performer all round – it outresolved the 50MP FX cameras wide open in the middle, and holds at the edges if you stop down a little. Very simple to use shift mechanism and rotation; it’s impossible to accidentally tilt because there’s simply no provision for it. A shame they didn’t make it 28mm, but honestly – it’s so good I force myself to make do with 35mm. Competitive even with the modern PCE and TSE designs, but sadly has the wide mount that’s not convertible to Nikon F.

Continued in part II.

__________________

Visit the Teaching Store to up your photographic game – including workshop and Photoshop Workflow videos and the customized Email School of Photography. You can also support the site by purchasing from B&H and Amazon – thanks!

We are also on Facebook and there is a curated reader Flickr pool.

Filed under: Gearhead: Reviews

Continued from part 1.

Remember: what constitutes ‘ultimate’ is actually highly subjective; some value smooth drawing quality and tonal transitions over outright resolution; others require zero distortion or high color accuracy or secondary color correction. If anything, my personal preferences tend to lean towards resolution and microcontrast; I can accept some vignetting, distortion, secondary lateral CA (but not longitudinal) – because these are easy to fix in post. Field curvature, smearing, coma etc. are not. Not all lenses on this list are here because of technical perfection or MTF chart performance, either. On top of that, there are two lenses that are not system options, but included anyway because they deserve honourable mentions. There are probably also better lenses I’ve not used yet (and so obviously can’t include them). I’ve tried to give justifications where possible. With that in mind, and in no particular order, here we go.

**Items denoted with two stars are lenses I currently own. *One star, lenses I’ve owned. Links are to reviews or affiliate suppliers. Images shot with the respective lenses mentioned.

HCD 28, H5D-50c

The Hasselblad HCD 28/4** (review B&H)
Though its sibling 24/4.8 is equally good – if not slightly better optically – I prefer the rendering of this lens, the angle of view, and the slight bit of extra speed you get. It’s crispy, transparent and has a very neutral personality with just enough everything to be transparent, but not overly dominating.

HC 50 II, H5D-50c

The Hasselblad HC 50/3.5 II** (review B&H)
In many ways, this is the ‘medium format Otus’ – no obvious weak points in the field or aperture, almost flat field, and excellent color correction and micro contrast. If I had to work with only one lens on the Hasselblad, I’d be torn between this and the next one…

HC 100, H5D-50c

The Hasselblad HC 100/2.2** (review B&H)
This lens continues to be a bit of a paradox to me. It’s both optically rather weak wide open, but oddly compelling: most of my favourite images (and those of the audience) shot on the H system were made with it; though I have to say my affinity towards it has increased since having my copy focus tuned so it locks on perfectly wide open. I think it’s because it has a very smooth rendering style and out of focus areas – both foreground and background. Microcontrast doesn’t come on song until about f4, and below that the lens is smooth rather than biting. At f4-8, it’s as sharp as you’d want – everywhere. I tend to think of it as a portrait or cinematic lens at f2.8 and below, or an excellent f4 lens with an f2.2 emergency speed. It also performs very well with the HTS and extension tubes for macro work. Another chameleon…

HC 150N, H5D-50c

The Hasselblad HC 150/3.2 N** (review B&H)
There’s something about the 150mm focal length on 44x33mm sensors; this is the second one on the list. It has a nicely compressed perspective that’s not too extreme, but at the same time more than enough to split a scene into planes. This lens has a high degree of secondary color correction and great bite at the focal plane, at all apertures. It’s one of my favourite documentary lenses especially when you can’t physically get any closer, but want to retain some intimacy in the scene. Originally, I borrowed this one for a job ‘just in case’, but landed up buying it because it landed up becoming one of the core lenses…

250 C Superachromat, H5D-50c

The Hasselblad Zeiss 5.6/250 Sonnar Superachromat**
Built in three versions, and all with the same optical formula. I have a very early first generation ‘C’ type; it’s one of the few lenses whose performance doesn’t appear to improve as you stop down, because it’s already so high at maximum aperture. It doesn’t have the macrocontrast of other Zeiss lenses, but microcontrast in abundance. Note there is no T* coating on any version of this lens, because it interfered with IR/UV transmission. Some care must be taken with flare. It’s tricky to use because of length and vibration and focus throw (very long, so transitions are slow) – but when everything lines up, the results are uniquely transparent – especially on the 50MP MF CMOS sensors. The medium format equivalent of the Voigtlander 180 APO-Lanthar.

CF150, 501CM, CFV-39

The Hasselblad Zeiss CF 4/150 Sonnar**
An underrated bargain. Best performance occurs stopped down so all secondary color has disappeared, but the transitions are so smooth, the layering so subtle and the price so low (typically $350-500 on the secondary market, depending on condition) that you can forgive it for effectively being a f5.6-11 lens. Yes, the 4/180 is better, but it’s also quite a lot more expensive, and significantly heavier. I carry the 150, but not the 180. And a lens not carried is a lens not used (and a picture left behind).

XCD 90, X1D-50c

The Hasselblad XCD 90/3.2 (B&H)
The MTF charts promised something spectacular, and boy did they deliver – the 90mm is the undoubted highlight of the native lenses for the X1D; it’s sharp, contrasty, detailed, and has very smooth foreground and background OOF areas. Performance is already outstanding wide open, and holds til you hit diffraction at f13 or so.

A150, 645Z

The Pentax 645 A 150/3.5*
A legacy lens from the manual focus days, but with some rudimentary electronics to log EXIF data. It’s another one of those truly tiny lenses – smaller than the Voigtlander 180 APO even – but a stellar performer when stopped down a little, and with truly smooth mechanics that make manual focus very easy. Again lacks the outright punch of the modern lenses, but is tonally smoother than Pentax’s later designs.

21 Loxia, A7RII

The Zeiss FE 2.8/21 Loxia* (review B&H)
This one surpasses even the ZF or ZE 21mm in performance, despite being a 52mm front thread lens, and overall tiny. It’s probably the best FF ultra wide out there, especially for its edge performance. The short flange distance of course helps. Too bad there’s no version of this lens in another mount – nor can there be.

The Zeiss FE 2/50 Loxia* (B&H Amazon)
There are now no fewer than three native FE mount Zeiss 50mm-e lenses; the 55/1.8 is a bit clinical and harsh, and suffers the typical Sony-made sample variation; the 50/1.4 is truly massive, but the 2/50 Loxia is a little gem and has a really pleasant rendering that seems to be very much at home with human subjects.

20 Art, D5500

The Sigma 20/1.4 Art** (review B&H Amazon)
Best bang for the buck wide angle, and there’s nothing faster. I think of this as a really excellent f2.8 lens, but with an f1.4 emergency speed that’s critical sharp over the middle third or so of the frame, even on the D810. It’s big and heavy and won’t take filters, but show me another lens that can do what it can…

The Sigma 18-35/1.8 Art* (B&H Amazon)
A real surprise because its performance is the equivalent of the Art primes in the same focal length range on DX (this is a DX only lens) – but it’s a zoom. In fact, I found it superior to the 35/1.4 Art. As a bonus, it covers 1.2x crop from ~20mm upwards with some correctable vignetting, and FX from ~28mm upwards. Once properly tuned with the USB dock, focus performance was fast, consistent and sharpness excellent. A bit large and the zoom ring was a bit stiff overall, but these are minor niggles for a lens of this performance (let alone this price and performance).

Leica Q

Honorable mention: The Leica 28/1.7 Summilux in the Q 116** (review B&H)
I throw this one in because it’s very much a modern lens in that central sharpness is always outstanding, with excellent edge performance – even wide open. Stopped down, it gets much better; however, it requires significant software correction to reach this level. We don’t normally notice it, so I’m prepared to give it a pass. The rendering style is very pleasant, and out of focus areas are surprisingly smooth and well-separated despite being a wide angle lens.

GR (V)

Honorable mention: The Ricoh 18/2.8 in the GR (V)* and GR II (VI) cameras (review B&H Amazon)
Yes, it’s a small format, yes, it’s a fixed prime – but for a lens of that size, that close to the sensor (the rear elements are actually larger than the front ones to spread out the rays for telecentricity and to avoid edge issues) – the performance needs no excuses. It resolves very highly at all apertures and in the corners, and surprisingly, distortion and secondary color are very minimal. Micro contrast is also excellent.

__________________

We are also on Facebook and there is a curated reader Flickr pool.

Filed under: Gearhead: Reviews

85mm lenses and equivalents on native or adapted formats – yes, I probably have too many. Upper left row: Nikon 85 PCE Macro, Zeiss 1.4/85 Otus, Nikon 24-120/4 VR, Hasselblad HC 2.2/100; middle row: Zeiss 1.4/85 Milvus, Canon EF-S 18-55 STM (APS-C), Nikon 85/1.8 G, C/Y Zeiss 2.8/85 Leitax converted to Nikon mount; lower right row: Zeiss Hasselblad CF 2.8/80, Zeiss Hasselblad C 2.8/80 T*. I wanted to add the Hasselblad HC 35-90 zoom, but it wouldn’t fit in the picture. And there also used to be a Zeiss 1.8/85 Batis, Zeiss ZM 4/85, Nikon 80-400 G VR and Voigtlander 90/3.5 APO, but I’m recovering now…

Though this post may seem like a hoarders’ justification more than anything – I can assure you, it isn’t. Whilst you could probably pick one lens in each focal length or angle of view and hack your way into making it work, there are some pretty solid reasons why you might not want to – and this is something I’d like to discuss today. Trust me, there are reasons why I’d prefer not to have to carry two or three seemingly overlapping lenses on assignment – but often there’s simply no choice. Here’s my logic, using the 85mm-equivalent focal length as an example.

Tricky shooting conditions: low light, and mostly monochromatic. Wide open, any poorly corrected lens will display serious aberrations.

Outright performance
This is the reason you’d choose an 85 Otus over the 85mm mark on your zoom, for instance. If you must shoot in very low light and therefore probably wide open, on a high resolution body, you probably the best lens you can get. Or you may be looking to extract that performance under ideal conditions – ideal apertures, fast shutter, 100% controlled lighting. This level of performance of course comes with tradeoffs in size, weight and portability; not to mention criticality of focusing before you start to lose all of that hard-won resolving power. It’s for this reason that when I have the opportunity, I’ll default to manual focus and live view: all AF areas are of a finite size, which means that they will almost always cover a subject that occupies a range of different distances. This variation in distance may be great or small, but the higher the resolution the sensor, the more critical focus becomes – and the plane which the camera chooses to focus on is not always going to be your intended one. For obvious reasons, this is obviously not practical with moving subjects.

A fleeting moment, impossible without AF.

Autofocus
The corollary of the previous statement: when I’m shooting fast-moving documentary work, especially under lower light conditions, AF is a must to have anything like a respectable keeper rate – especially if there is no opportunity to reshoot the image. I use the 85/1.8 G here because it focuses quickly and accurately, and performs well wide open (though obviously not as well as the 85 Otus). For those of you wondering why I don’t use the f1.4 G, it’s to do with performance tradeoffs: the lateral chromatic aberration at maximum aperture is quite visible and distracting, and if I have to use the lens at f2 or smaller anyway to divest myself of it, there’s not a lot of point in paying the significant premium or carrying the extra weight (plus f1.4 is only 2/3 stop faster than f1.8 anyway). There is not much difference in AF speed, either. For similar reasons, if I’m working under bright conditions or mostly stopped down, I’ll use the 24-120 zoom.

Weight
None of us like to carry more than we have to, but there’s always a tradeoff between image quality and overall size/weight. Where that diminishing returns curve kicks in depends on the individual, of course. However: if I’m not going to need wide apertures, or AF, I like the C/Y Zeiss 2.8/85: nice rendering, fast enough, and absolutely tiny. Plus, if you’re using it at f8 or smaller for landscape work – there’s almost no difference between that and an Otus anyway. If you replace all of your 1kg f1.4 primes with 250g f2.8 primes, that’s enough weight saving for a very decent tripod. And I know which will open my shooting envelope wider and make the bigger difference to image quality…

Tilted focal plane to render the entire watch back in focus – single shot, no stacking.

Special purpose needs
I think this category is pretty self-explanatory: you’re not going to produce very good macro images with a lens that cannot focus closer than 1m, even if you use extension tubes. The optics simply aren’t optimised for close range reproduction (the exception perhaps being the 85 Otus, owing to its degree of correction – with extension tubes, it does perform nearly as well as the 85 PCE Macro even close up). Even so, you might get away with this at lower resolutions or output sizes. However, you can’t cheat extended depth of field or perspective control: tilt and shift are tilt and shift. It is for this reason that the 85 PCE is the only real choice for this kind of work unless you want to focus stack, which might not be possible in some situations.

Difficult to demonstrate this one at web sizes; what I’d like you to pay attention to here is the smoothness of out of focus areas.

Specific rendering characteristics
Perhaps the best example of this isn’t an 85mm; it’s the (in)famous f1.0 Leica 50 Noctilux with the swirly bokeh. It’s undeniably distinctive, but this comes with a related problem: it’s too distinctive. Everybody else who owns this lens will also shoot it wide open (who pays a premium for a f1.0 lens to use it stopped down when there are better alternatives?) and land up with images that look the same. This is perhaps an extreme example, of course. Let’s take the two Zeiss 85mm lenses: the Otus 85 delivers ultimate resolution at all apertures but at the expense of smoothness of background out of focus areas under some circumstances because of the aspherical element; the ‘bokeh balls’ are sometimes seen to have texture. The aspherics are part of the optical formula required for that level of apochromatic correction. On the other hand, the Milvus 85 has no aspherics, and a hint of coma and both lateral and longitudinal chromatic aberration wide open, but extremely smooth transitions and out of focus areas. The CA on the Milvus is mostly gone by f2.8, but the bokeh ball texture on the Otus remains. This difference can be quite visible in some kinds of portraiture, for example – which is why I keep the Milvus around.

Not quite 85mm: 75 was required, 75 was available (but not with a prime.)

Flexibility
There are times when you need just a bit more or just a bit less – and you can’t physically move, either through a lack of anywhere to go or a consequent change in perspective and foreground-background relationship. Or, you may simply not want to change lenses given ambient environmental conditions – the only thing for it is a zoom, of course. The highly polarising 24-120/4 is my choice here: it covers just about everything, is fast enough, good enough quality-wise, and pretty much all you need so long as you have enough ambient light. (Of course, your mileage may vary if prefer shallower depth of field; I work stopped down most of the time so it makes no difference to me.) Sometimes, those edges can make or break an image – it’s nice to have the choice. Just don’t get lazy and start making huge compositional and perspective changes with the zoom ring rather than your feet…

100mm on a 44×33 sensor is roughly 79mm – close enough…

Different formats
I think this is perhaps the most contentious of all points here, hence my leaving it for last. The most obvious reason is that we may want more or less DOF for a given angle of view and aperture (and be unable to stop down or open up more for various reasons) – thus necessitating a change in format to achieve the desired presentation. More complex is that though a given angle of view should render foreground-background relationships and projection identically for an ideal lens, the truth is that not all lenses project or render ideally, which is why we have the various kinds of distortion. In practical terms, it is very difficult to make a wide angle that has a flat plane projection; there’s invariably some barrel distortion that changes the relative proportions of the subjects and the way they appear to us. It may be a subtle difference of a couple of percent at most, but it’s still noticeable – even if we don’t consciously pick up on it. It is for this reason that most of the time, a 10 or 14mm lens on a smaller format will not look the same as say a 28mm lens on medium format. Coupled with this, we have depth of field considerations: even though we may have the same angle of view, 10/2.8 won’t render the same as 28/2.8 because of the difference in real depth of field. Now, here comes the contentious bit: due to optical design limitations*, in general, the larger the format, the less distorted the rendering** – even if you have the same overall angle of view.

*Part physics, part system related, part ‘what-can-you-do-at-a-given-price-point’ – people buying a small sensor probably aren’t willing to pay for or carry a perfect 12mm, though it could of course be made.
**By a similar token, if you’re paying for medium format, you’re probably going to be willing to pay for and carry that perfect 28mm.

I hope at the end of this you can see why lens selection at a given angle of view isn’t a straightforward thing, even if you’ve only got one mount or system. There may not be that many native choices for some systems, but adaptors complicate the mix significantly: especially those that allow movements when stepping down from a larger image circle to a smaller one (e.g. medium format to 35mm, or 35mm or APS-C or M4/3). Just beware of planarity issues as always, though the compromise might be worth it for infrequently encountered situations. It might not be worthwhile investing in a tilt shift macro if you don’t shoot off a tripod very often, or require high magnifications, for instance. But at the same time, a zoom might not cut it you do a lot of portraiture. I really think the only thing you can do is identify the most common kinds of shooting situations you encounter, and work backwards from there in the most commonly used focal lengths (shooting for a while with a zoom and then doing a little EXIF analysis might help here). MT

Some of the lenses mentioned in this post are available from B&H or Amazon:
Nikon AFS 24-120/4 VR** – review B&H Amazon
Nikon AFS 85/1.8 G** – review B&H Amazon
Nikon PCE 85/2.8 Micro** – B&H Amazon
Zeiss ZF.2 1.4/85 Otus APO-Planar** review B&H
Hasselblad HC 2.2/100** – B&H

__________________

We are also on Facebook and there is a curated reader Flickr pool.

Filed under: Articles, Gearhead: Reviews, On Photography

After my review of the Olympus M.Zuiko 17mm F1.2 PRO, a few people suggested a lower priced yet seemingly competent alternative – the Sigma 16mm F1.4 DC DN Contemporary lens. A dear friend, Amir, who recently obtained a Sigma 16mm was kind enough to loan it to me for review purposes. So is this really a budget-friendly option to the 17mm F1.2 PRO from Olympus, and does the Sigma lens perform well enough under the standard Robin Wong lens torture tests?

Some quick disclaimers; neither Ming Thein nor I are associated with Sigma Malaysia. This is an independent review and my approach is always based on user experience and may be subjective. The Sigma 16mm F1.4 was used on the Olympus OM-D E-M1 Mark II for all the sample images shown in this article. All images were shot in RAW and post-processed using Capture One Pro, with minor adjustments. You may view the images from this article in higher resolution on an online Google Photo album here.

F1.4, 1/40, ISO320

F1.4, 1/25, ISO500

F1.4, 1/2, ISO200

The Sigma 16mm F1.4 lens was originally designed to fit the Sony APS-C E-Mount but has since been adapted to a Micro Four Thirds mount. That explains why it is slightly larger and heavier than the Olympus 17mm F1.2 PRO lens. The Sigma 16mm lens has a filter size of 67mm and weighs about 405g.

Sigma has been highly praised for their newer line of lenses available to both DSLR and mirrorless system camps and built a reputation for delivering high performing optics at reasonable price points. The Sigma 16mm F1.4 lens construction consists of 16 elements in 13 groups, including 3 FLD, 2 SLD and 2 aspherical elements. The lens also features a non-rotating front element which is good news for filter users, and a respectable minimum focusing distance of 25cm (with magnification of 0.1x). Also worth noting that the Sigma 16mm is sealed from water splash and dust.

For a full specification list, check out the official product page here.

The 32mm equivalent focal length is an odd one to work with – it’s in the middle of the more traditional 28mm and 35mm focal lengths. Being wider than 35mm allowed me to use the Sigma 16mm lens as a general wide angle lens in my test shooting sessions. What should I do with a wide angle lens with a F1.4 aperture? Take it for a stroll at night on the streets of Kuala Lumpur and shoot the dimly lit scenes of the urban landscape. After all, a F1.4 lens allows more flexibility in low light conditions.

F2.8, 1 second, ISO200

F1.4 (left) vs F2.8 (right), crop from previous image

F1.4, 1/8, ISO200

The Sigma 16mm F1.4 performed favorably well under dim street lights. Autofocus (when used on the E-M1 Mark II) was snappy and reliable, I found no issues locking focus even in extremely dark situations. It was a joy to have a F1.4 lens, coupled with the capable Olympus 5-Axis Image Stabilization – I could get away with hand-held shots at ISO200 most of the time. This is the largest benefit of shooting with Micro Four Thirds: I can get away with perfectly sharp images at 16mm with a 1 sec shutter speed at ISO200. With a few fast prime lenses, there really is no excuse why Micro Four Thirds shooters can’t shoot in the dark.

As I inspected the images from this short session at night, I found the wide open images to be uncomfortably soft. They were not unusably soft, just not as sharp as I was expecting. I then decided to do a side by side comparison by stopping down the images to F2.8 to see the difference. As shown in the center crop of a building image above, the F2.8 crop displays significantly more detail, contrast and higher clarity than F1.4. Bear in mind that the F1.4 image was shot at 1/5sec shutter speed, while the F2.8 image was taken at 1 second shutter speed, both hand-held. Since there was a chance that the image stabilization was not optimized for the Sigma lens (after all, it is a third party lens), I conducted a more thorough test on sharpness at different aperture values in a daylight situation.

F1.4, 1/250, ISO200

F1.4 (left), F1.8 (center), F2.8 (right) Bokeh Comparison

F1.4, 1/640, ISO200

Crop from previous image

Under better light, it’s clear that even wide open, the Sigma 16mm does produce decently sharp images, but nothing to write home about. At best, the sharpness is close to what the usual kit lenses from Olympus and Panasonic can do. Notably, Olympus and Panasonic prime lenses perform considerably better wide open and rarely require stopping down.

The bokeh rendered by the Sigma 16mm F1.4 lens is fantastic. When you can get close enough to your subject, shooting at wide apertures allows a good amount of shallow depth of field. The background just melts into creaminess and I really like what Sigma is doing with the out of focus areas of the images. This is no “feathered bokeh” but to be completely honest I can’t tell you what feathered bokeh is. All I can say is, I know good bokeh when I see it and the Sigma 16mm produces buttery smooth and pleasing looking bokeh.

F4, 1/2500, ISO200

F1.4 (left) vs F4 (right) Click for larger view, crop from previous image

F4, 1/3200, ISO200

F1.4 (left) vs F4 (right), corner crops from previous image

F5.6, 1/15, ISO200

To achieve optimum lens performance, you need to stop down to at least F2.8, or even F4. The difference between shooting at F1.4 and F4 is huge, as shown in the building crops above. I am not expecting the Sigma to be super sharp at F1.4, but I was also hoping it could deliver a bit more punch, because lets face it, if I am buying a F1.4 lens, I don’t want to be shooting it at F4 all the time.

Chromatic aberration was quite well controlled to my surprise. Honestly I have seen worse purple fringing on Panasonic lenses (even on some of the high end ones). This could be due to automatic correction profile. Chromatic aberration is more prominent at high contrast areas especially shooting wide open (though still perfectly tolerable), but as the lens is stopped down the chromatic aberration slowly diminishes.

Again, thanks to modern software, there is no trace of barrel distortion, and straight lines appear perfectly straight in my shots. I do not mind if the optics are not perfectly corrected, as long as I don’t see any ugly curvature.

F4.5, 1/50, ISO200

F2.8, 1/100, ISO200

From top left, then clockwise: F1.4, F1.8, F2.8, F4 – Bokeh comparison

Minimum focusing distance, the nearest

F5. 1/100, ISO200

I was actually quite satisfied with the close up shooting capability of the Sigma 16mm lens. While this is no macro lens, it does go really close – the above samples of shooting the Green Arrow Lego mini-figure was actually at the minimum focusing distance. At such short distances, the lens delivers crisp and highly detailed results.

The Sigma 16mm F1.4 lens does a great job combating flare. Certainly I have seen worse flare control in some Panasonic and Olympus lenses. I shot directly against the sun, bright spot-lights and all kinds of backlight situation and I almost never encountered any flare or ghosting problems.

Does the Sigma lens feel a little too big and heavy to use? This is a tricky question to answer – for a Micro Four Thirds lens, the Sigma 16mm F1.4 is larger than most of the prime lenses available for the system, even larger than the Olympus PRO primes which already received many complaints for being too big. Having used the lens on the E-M1 Mark II for a couple of days now, I honestly never felt that the lens was uncomfortable in my hands or was too large at any time. I felt the size and weight was well balanced when used with E-M1 Mark II. However, I cannot say the same if you attach the Sigma lens on a smaller Olympus body, say the E-M10 series, or PEN cameras.

I had my doubts about the AF performance being as good as the native Olympus and Panasonic lenses. The older Sigma lenses for Micro Four Thirds have had slower and more hesitant AF, but the new Sigma 16mm F1.4 performed flawlessly on the E-M1 Mark II. Focusing was just as quick and reliable as any other native lens I have used before.

F16, 1/3200, ISO200

F3.5, 1/30, ISO200

F1.4, 1/50, ISO200

On the whole, I do think that the Sigma 16mm F1.4 DC DN Contemporary lens is a good match for certain Olympus or Panasonic Micro Four Thirds bodies – as long as you remember the less than stellar optical performance at F1.4. While the lens is a little soft wide open, when stopped down it is just as good as any of the higher grade lenses from native Olympus and Panasonic offerings. The Sigma renders excellent bokeh and does a great job shooting close up. Chromatic aberration, distortion and flare are all well controlled (perhaps with some software correction help). Autofocus works well with the OM-Ds and there really is not much to complain about the lens. Considering the much lower price point against the Olympus 17mm F1.2 PRO, the only compromise is the lackluster performance when wide open.

I will be using the Sigma 16mm F1.4 lens for a few more shutter therapy sessions, so you can expect more sample images soon.

If you own the Sigma 16mm F1.4 lens for Micro Four Thirds, I would love to hear your thoughts. Do share your experience!

The Sigma 16mm F1.4 Dc Dn Contemporary Lens is available from B&H

The Olympus OM-D E-M1 Mark II Is available from B&H

__________________

We are also on Facebook and there is a curated reader Flickr pool.

Most of you know me for being at the bleeding edge of hardware and being able to deploy the difference – that was true at least until my back injury last year, which has severely limited what I’m able to carry for any length of time. It has forced me to look at things I would normally have ignored; for whatever reason, in this industry light and small is usually also synonymous for ‘entry level’ and ‘cheap’. But in doing so, I’ve found some surprising hidden gems: hardware that most people pass over at face value for lack of bragging rights or seemingly ‘obvious’ deficiencies. Be prepared to be surprised, I was. This will be the first in a series of el cheapo reviews.

When I started off with DSLRs, the king of Nikon wides – DX only at the time, of course – was the AF-S 12-24/4. It was a decent performer even on the 12MP bodies, but started to fall apart with anything much more resolving than that. Distortion was…spectacular and not easily correctable. It was also very much a prosumer build lens, with light plastic everything, the slower AF-S motor and a non-prosumer whopping $1200 price tag or thereabouts. Fast forward fifteen years and we now have a successor (there was also the AF-S 10-24mm f3.5-4.5 DX, which sits somewhere between the two in price, build and optics). It takes many things away: it isn’t as long (20mm vs 24mm); it has an even plastickier build – even the mount is plastic – and it’s pretty much a stop slower across most of the range. BUT: it is $280 or so, new, from your choice of online outlet, and has VR, the new fast AF-P pulse motor, and weighs just 230g. It even covers FX from 13mm upwards, though the Z6 and Z7 will auto-crop to DX and you can’t override this. If you are not a fan of long reviews, then just enjoy the images, skim through the rest, and click the buy link at the bottom of the post.

To put things in context: this lens cost the same as my camera (an albeit unusually cheap D3500 kit) but less than a medium format battery. It continues to amaze me how far optical design is being pushed at both ends of the spectrum: we have some truly amazing premium lenses, but by relaxing one or two of the usual design constraints (maximum aperture, distortion, secondary corrections, use of moulded plastic asphericals etc.) we also have some truly amazing budget lenses.

I suspect the rationale goes like this: the company has already bought or developed all of that expensive optical computation software; the price-volume curve at this end of the market is wildly nonlinear, and something like the 10-20 is probably going to sell two or three orders of magnitude more lenses than say the superteles or PC-Es, and be more profitable as a total project. More resources than you might expect get thrown into it, and suddenly – you actually have a really good lens.

Don’t get me wrong: Nikon has clearly cut corners here. The mount is plastic; there are no switches for AF/MF or VR on/off (that has to be done in a menu, and some cameras require a firmware update). The AF-P motor is only compatible with the most recent generation or two of cameras. There is of course no rear gasket. Hell, it’s so cheap that the regulatory labels and MADE IN THAILAND are part of the injection moulding of the shell – they aren’t even a separate sticker. Speaking of which, the nameplate – metal on even Nikon’s midrange lenses – is a sticker here. Even the mount indexing dot isn’t the usual glued-in plastic ball: it’s printed.

Like the rest of the DX AF-P lenses, it’s a class leading study in both marketing stinginess and engineering brilliance: despite all of the compromises, the lens feels neither hollow, nor insubstantial nor flimsy. The zoom ring rotates with constant friction and good smoothness, and none of the usual plastic-on-plastic stickiness; it doesn’t feel very different to my gold ring Nikkors, to be honest. And the focus ring is fly by wire, but silky smooth and very nicely calibrated.

Where Nikon hasn’t cut too many corners is in the stuff that really matters: optics and mechanics. The 10-20 uses a surprisingly complex 14 elements in 11 groups, with three asphericals and (surprisingly) no ED elements. There are however a lot of achromatic doublets used in the design: three or four, depending on how you’re counting. Undoubtedly this keeps unwanted dispersion to a minimum. There is very little longitudinal chromatic aberration but some lateral visible at wider focal lengths; the small aperture probably helps too. (It also means that peak resolution is found somewhere between maximum aperture and f8, after which diffraction starts to soften things a bit on the 24MP DX bodies.)

The optical tradeoff made appears to be in field flatness and distortion. There’s some strange focal plane curvature going on towards the corners that’s more visible at close distances and seems to go away a bit at infinity. This manifests as softer corners, but actual behaviour seems to be more like strong rearward field curvature. It’s even clearer when you put the lens on a FX body: there’s a really abrupt dropoff in the corners of the image circle beyond DX. So although you can use it from 13mm upwards on FX, you probably won’t want to because the corners take a beating. In fact, the image circle around 17mm or so is actually larger than at 20mm. However: if you have a D850 and don’t mind using the 1.2x crop, it’s actually pretty good: you have a very cheap wideangle that will cover ~15mm-equivalent upwards, and still keeping better-than-DX-resolution (about 30MP or so).

The actual characteristics of distortion are also worth some discussion: it’s mostly simple, of the barrel-type, at every focal length. There is almost none of the messy moustache-type (uncorrectable). It also avoids the typical wide angle projection* distortion: the edges don’t get stretched excessively larger compared to the middle, so the images appear surprisingly natural in perspective for a lens this wide. (I would still of course recommend holding the camera level.) This appears to have been one of the main design tradeoffs: distortion against field curvature and sharpness. It also isn’t entirely symmetric on my copy, which suggests some sensitivity to sample variation; I didn’t have a second copy to test.

*Wide angle foreground-background emphasis is NOT the same as projection: projection is when you flatten 3D space into 2D, e.g. for maps of the earth. Frequent stitchers will know there are a lot of different ways of doing this and the results can be very different: but what is stitching other than creating an effective angle of view wider than the one you’ve got on the camera?

Flare is very well controlled, with only one or two purple ghosts at the wider focal lengths. Coma is similarly well controlled, with some smearing visible in the corners on the 24MP DX bodies, but nowhere near as bad as you’d expect. If you’re wondering why I haven’t talked about sharpness/resolution and bokeh – it’s because there’s nothing much to report. Central sharpness is as good as you can wish at pretty much all focal lengths from wide open, but falls off into the corners due to the previously mentioned field curvature. As with all recent Nikon lenses – macrocontrast is high, with microcontrast probably best described as coarse. It isn’t as refined as the Z lenses or gold rings, but it also doesn’t have the same coatings or price tag. Finally, don’t even bother talking about shallow DOF: remember this is a slow superwide.

I’m going to conclude by stating the obvious: Nikon has made an impressive lens, made truly impressive by the price and size. The tradeoffs are sensible and can easily be worked around. For the DX shooter, it’s a no-brainer: buy one. For the FX shooter who only needs the occasional wide: it’s also a no-brainer at the price: buy one. For the serious wide shooter, you probably already have your poison of choice – but if you want to go light and still have the wide option – you guessed it, buy one (and a D3500). And yes, falling into the second camp, I’ll be picking up one of my own. MT

The Nikon AF-P 10-20mm f4.5-5.6 DX VR is available here from B&H.

__________________

Ultraprints from this series are available on request here

__________________

Visit the Teaching Store to up your photographic game – including workshop videos, and the individual Email School of Photography. You can also support the site by purchasing from B&H and Amazon – thanks!

We are also on Facebook and there is a curated reader Flickr pool.

Revisited and reviewed: The Zeiss ZF.2 2/100 Makro-Planar T*

Understanding lens-related descriptive terms

Review: The Carl Zeiss ZF.2 2.8/21 Distagon T*

Lens review: The Nikon AF-S 85/1.8 G

Lens review: The Nikon AFS 60/2.8 G Micro

Lens review: The Leica 35/1.4 Summilux-M ASPH FLE

Lens review: The Nikon AF-S 80-400/4.5-5.6 G ED VR II N

Lens review: The Panasonic Lumix Vario PZ 14-42/3.5-5.6 X G

Lens review: The Olympus 12-40/2.8 M.Zuiko PRO

Photoessay-review: the Nikon AFS 70-200/4 VR and Havana cityscapes, part I

Lens review: The Zeiss ZF.2 1.4/85 Otus APO-Planar

A visit to Zeiss and thoughts on the Milvus line

Review: the Zeiss 1.4/28 Otus APO-Distagon

The ultimate lens list, at Nov 2016 (part I)

The ultimate lens list, at Nov 2016 (part II)

Reasons to have multiple lenses in the same focal length/AOV

Review: the Sigma 16mm f1.4 DC DN C

Wider please, but on a budget: the Nikon AF-P 10-20mm f4.5-5.6 DX VR review