Silly me, I approached this review of the Zeiss Milvus 85 all wrong!
Excited by the prospect of comparing the newcomer to my almighty OTUS 85, codename Hubert, I devised a series of test to establish how close the Milvus comes to the performance of its older superstar brother. A very dualistic worldview, which shames me a little.
Time, then, to reboot this and find more interesting things to narrate. Oh, the comparison will be published soon, and – spoiler alert – it is not a clear cut affair. However, for the time being, why not focus on determining whether this Milvus, by itself, belongs to the select club of über-lenses instead?
And, that end, is it time to finally explain our codename system and out classification of lenses, maybe?
There are a vast majority of soulless lenses out there. The zooms designed to reduce the lab-tested flaws to a minimum at the expense of doing nothing brilliantly, for instance. Much like any commoditized product and the communications of so many brands, blandness seems a safer choice than flawed excellence. We don’t have a name for these lenses at DS, because we’re simply not interested in reviewing them. Life. Short.
Then, there are interesting lenses that fall short on some technical criteria. Think C-Sonnar 1.5/50 ZM, absolutely brilliant in some conditions, and slightly oh-hum out of its comfort zone. Or my beloved Distagon 1.4/35 ZM, probably the best 35 lens ever designed for 90% of its perimeter but unhappy in the corners of a thick-stack Sony sensor.
We have pet names for these lenses, meant to highlight the character of the lens. Audrey (Hepburn), for absolute delicacy, beauty and mischief, is the Distagon 1.4/35 ZM. Cesar is the C-Sonnar, short, mighty and ill-tempered.
The OTUS 85, we call Hubert, because that sounds like über (in French) and that lens defines the essence of optical magnificence. Free of any sort of real-life nasties how ever you push it (yeah, you can find a few purple pixels here and there is you know how to light up the scene, flare is not totally absent and coma is quite plainly evident in the corners in astrophotographic conditions. We’re talking real-life photography niggles here, and the OTUS 85 simply doesn’t even understand the concept).
But beyond that, it has a way of collapsing the universe into sharp/gaussian unsharp and serving-up gargantuan doses of 3D that nothing I’ve ever used even comes close. See below (the only non Milvus 85 photograph on this page).
So our work is cut out for us.
Let’s not get into the technicalities of the Zeiss Milvus 85/1.4, you can find all of that on the official datasheet. But it is worth remembering that this new lens takes the highly unusual fully-spherical design-path (compare that to the 9 aspherics in the Batis 25 from the same stable) and that its MTF curves at f/4 indicate the highest levels of brilliance right up to the very edge of the frame (though not quite into the corners).
The corners weren’t the first concern of the design team, given that the portrait-oriented tog used as design persona. And the following 3 photographs (made by myself, a definitely non-portrait-oriented photographer) serve to prove how well the team succeeded at this main goal.
But few of us, in the Zeiss marketing funnel, are going to spend 2 grand on a portrait-only single-trick pony. So how well does this Milvus 85 perform in other photographic scenarios? Is it a good B&W lens? Is it good for landscape? Can I use it on the street? Does it hold its own in architecture?
And, most-importantly, how neutral is the lens? We’ve all drooled over the capuccino look of a Noctilux. But, after a while, every Notcilux picture looks like a Notcilux picture. The lens very quickly dominates the look of the image. Which is OK for me when I’m buying a 200$ LTM, but not for a 10k behemoth. Cheap digital filters are fine for quick fixes but when I overspend on a lens it is to ensure it excels at the widest possible range of use scenarios.
To find out how neutral a lens, the simplest method is to create a set of photographs and process them freely in different ways to determine whether the lens always takes you in the same direction or, on the contrary, sets you free to explore.
So, I have separated my photographs from a quick shoot in my home village into colour and b&w to find out for myself.
OK, all of this is subjective and labs will determine colour fidelity far better than my eye. But this lens may well produce the most pleasing colours I have ever seen. Not as subtle as Audrey (Distagon 1.4/35 ZM) but full-bodied, happy, non-diluted, differentiated, wholesome and true.
My previous post about the Milvus 85 already gave stunning examples of colour reproduction, but these were mostly made in the ultra-saturated wonderland of a Disney amusement park, mostly at night. The same can be said of the launch photographs we are seeing of the new Sony Master G lenses (which seem very appealing).
Real-life is different, though, full of subtle nuances and hard to reproduce hues.
So I took the lens for a stroll through my little hometown in mid-afternoon winter light on one of our signature cool and crisp mediterranean days, to see how well the lens coped with the faded colours of our narrow lanes in natural light.
And the lens doesn’t disappoint.
One tell-tale quality you can easily determine for yourself with your own lenses is that you do not want to touch the saturation slider in your post-processing software, or apply any filters, but simply alter the feel with contrast and luminosity adjustments.
The photograph above is a case in point. The marvelous dynamic range of the Sony A7rII made it possible to maintain detail in the shadows as well as in the brightest highlights of this scene, but the picture still looked a little dark and uninteresting. I increased contrast and bumped up the shadows, to a point where the orange on the corner wall almost seemed too saturated. So the saturation slider was indeed used, but to tone things down a little. And the result is both very colourful but also very believable.
The Milvus uses special glass to mitigate the inevitable effects of dispersion and keep colours clean, surface coatings are probably top-notch. But, mostly, this is a medium contrast lens that doesn’t blow every detail into unnatural proportions. There’s a huge amount of information in the files (very high-resolution) but the focus wasn’t placed on super strong micro-contrast. This leads to more relaxed aesthetics and to more pronounced mid-tone colours. Very high contrast lenses often require a lot of post-processing help to boost colours. Not so here. Laid back and colourful is what life looks like through this lens 🙂
So, all well and good when it comes to life-like and relaxed colours. But shouldn’t that easy-going character make the lens a slouch in b&w ?
I have no easy answer to that because most B&W work is heavily edited and typically reflects more on the quality of the camera than on the lens’ abilities. And the Sony A7rII totally cured my Leica Monochrom itch, so the B&W pics on this page are inevitably helped by that tremendous imaging tool.
Still, though. My approach to evaluating a lens for B&W work is to provide various styles of post-processing so that readers can find a style that matches theirs and compare the results to what their own system provides. And few should be disappointed.
So, again, here are some samples form my brief walk through my village and a variety of inspirations from dark Antonioni to Adams to Foujita to local newspaper (for those of you old enough to remember the newspaper).
What I’m seeing here is a personality almost entirely dictated by post-processing. The tendency is towards medium contrast. Compare this to the photographs made with a Loxia 35 just a few days before, just a few miles away. Clearly not the same imaging universe. But still the lens responds to all sorts of processing queues, from the very high contrast gentleman at the very top of this page, to the alarmingly precocious almond tree just above.
Bokeh, and the contiguous concept of 3D also play an important role in determining how splendid/uninspiring a lens ranks in our weird classification system. Personality is good, but distraction isn’t so I’m not overly fond of the swirly school of thought ot of high edge visibility in out of focus areas.
And on that score, I’m seeing an A, but not an A+.
On this most horrid of bokeh tests, the f/8 photograph feels a little agitated. Not because the bokeh is bad but because the main subject doesn’t benefit from the OTUS-like high contrast that ensures super crisp separation from the background, whatever the background looks like. Here competition between foreground contrast and background contrast isn’t the best I’ve seen.
At a more normal (for subject separation) f/1.4, things are a lot better. There does seem to be a small amount of swirl here, but I’ve not seen that manifest in many other photographs so it might just be a case of background subject matter.
In a non-studio portrait scenario (subject 1m50 – 2m away, busy background), things look rather good. At f/1.4, one could hope for a little more blurring of the background but subject separation is excellent. The concrete wall displays and odd stretching pattern reminiscent of the mini-swirl of the previous example. But the photo is technically pleasing.
F/2 might be the prettiest setting, with the background not that much more present and the oddities totally under control. At f/4 (bottom photo), some very minor stray colour on the dark tree (sphero-chromatisicm ?) is barely visible but the photograph is very clean and pleasant.
In this final duo, I have enhanced the visibility (luminosity + contrast) of our feline model and lowered the background highlights. In this situation, which reflects normal conditions with no background trap) the results are among the most pleasing I have ever seen. The cat doesn’t jump out of the screen as it would with an OTUS 85, but the sense of presence and depth is superb. Colours are rich and natural. Bokeh is perfect. Absolutely nothing get in the way of the little story being told. Abfab.
As you’ll have guessed, I like the Zeiss Milvus 1.4/85 a lot. Long-term readers know of my infatuation for Mandler-era Leica lenses. To my eyes, this Milvus is a well-behaved successor to these magnificent lenses. Every bit as appealing, but without any of the misbehaviour.
Compared to an OTUS, it provides a more “integrated” image, with less 3D pop and a more painterly style. Hopefully, we’ll be able to compare it to the Sony Master G of similar spec and price to determine how close or different the drawing styles of the two lenses are.
So, excellent performance all around and an absolute master achievement relative to price.
How can I put this?
My gut reaction would be to call this is a minor masterpiece, but not an über-lens. As niggle-free as one could realistically hope for (yeah, lab tests will trash it in the corners at infinity and full aperture) as it is, it somehow falls short on the second über-membership condition. It somehow fails to take my breath away with any spectaculars in the way the OTUS so masterfully achieves.
While it may not grab you in the first seconds of viewing, the Milvus manages to maintain your attention a long time because of the subtlety of its imaging. Take the fuselage on the above two-frame stitch, for instance. This could be a studio shot made with a medium format camera but it’s a hand-held pic grabbed in between passers-by. Absolutely remarkable texture, 3D and colours make this life-like and more lasting than instaneously grabbing.
The Milvus’ super-power, then, might be its über-delicacy. So, maybe it’s time we rethink out classification system. There has to be room for the non-spectacular in the elite of photographic lenses. If so, this Milvus certainly belongs to it.
Please log in again. The login page will open in a new tab. After logging in you can close it and return to this page.