Welcome to part 6 (the last part) of the JAPB comparison of nine fast 35 mm focal length lenses.

This is a big article, which has been subdivided into several parts for convenience and the sake of humane loading times. See table of contents below.
• Part 1: Introduction, the lenses: pedigree and handling
• Part 2: IQ-comparison I – The Brick wall test
• Part 3: IQ comparison II – Urban vistas
• Part 4: IQ-comparison III – Bokeh and feel
• Part 5: IQ-comparison IV – Night-time vistas
• Part 6: Summary and conclusions (you are here)

Summary and conclusions

Firstly, the point of a side-by side comparison is to be able to evaluate a lens’ performance in the most honest and brutal way: by looking at the performance of a lens and comparing it against the performance of another lens. Again, we’re not testing all samples of a lens ever made, but only the samples we have on hand (read more in JAPB principles for reviews and comparisons).

Now unless you want to read some more philosophical musings about what makes a good lens (next), you might want to jump ahead to the results.

What makes a good lens?

We believe that comparisons such as this can offer you a great deal of knowledge – it did for me. I’ve had all these lenses prior to subjecting the to this review. I’ve used them, taken pictures with them, and scrutinised the results. Still, only by scrutinising a photo side by side with another photo can I start seeing where each lens has its strengths and weaknesses.

Admittedly, some strengths and weaknesses are evident from looking at a lens’ specifications. Obviously a lens with a maximum aperture of f/1.4 is brighter than a lens with a maximum aperture of f/2.4; a lens with a shorter MFD is more versatile than one with a significantly longer MFD, and a heavier lens is less fun to schlepp around. That said, even while such characteristics may be evident without needing to do more than looking at a data sheet, the significance of such characteristics (such as the utility of a short MFD) are relatively individual.

On the other hand, there are other characteristics which are not discernible from a data sheet, such as whether the aperture ring offers intermediate stops, or how pleasing the aperture ring is to use. Among such traits are also focus ring motion, how easy the lens is to use without looking (where the control rings are located and whether they are easy to discern and grip), as well as the overall balance (centre of mass) of the lens. All these have an impact on the lens’ usability (even though usability also is dependent on individual needs) . On the other hand, these are all traits which you can ascertain without ever needing to take a single photo with the lens. Simultaneously, should the image quality be utter garbage, one could argue that whatever ergonomic affordances the lens might offer are utterly meaningless. If all you’re looking for is a paperweight, there are surely cheaper approaches. And who really needs paperweights in this day and age anyhow?

Not surprisingly therefore, evaluations of optics tend to place significant emphasis on the optical qualities of lenses. This comparison is no different. Image quality, however, is not a uni-dimensional concept. A lens might offer high resolution but lacklustre contrast performance; a lens’ performance in daylight may be stellar, while its low-light performance might suffer.

The problem with passing judgement

Problematically, we should not assume that all photographers share preferences. One might need a lens for nighttime street photography and another wants a lens for architecture photography, one may value bokeh over sharpness, one may want a legacy lens for adapting to digital while another one might need the lens for his equally old film SLR. In short, the problem with all judgements about a lens’ quality is that the judgements are personal – related both to the needs of that person as well as to that person’s points of reference. In short: YMMV.

Some review sites aim at reducing a lens’ performance to a percentage, or an average grade (maybe even with decimals) or a number of stars or turtles… All such approaches assume that the preferences and use cases of all photographers are identical. While making good copy, this approach is fundamentally flawed. Just to take one example: with the renaissance of analogue photography, distortions and vignetting (two lens handicaps easily corrected in digital post production) are again becoming important. Thus whether such handicaps should be of interest depends centrally on the intended uses of the lens. Hence, reviews should either limit themselves to enumerating and describing the traits of a lens or – should they aim to be prescriptive – sufficiently take into account the typical usages of the lens.

Results

Make no mistake, each lens evidenced both strengths and weaknesses – not that they all need be evident from the photos unless you are in the habit of making large-scale blow-ups or are a pixel peeper (as many photographers are today). Also, while some lenses displayed more weaknesses than others, all these lenses were very capable of producing decent results – when used correctly and when circumstances were not too adverse. The following paragraphs will try to briefly evaluate each of the tested lenses. Remember – again – that these results are only indicative of these, tested samples.

Canon FD 35 mm f/2 ‘concave’

Basic specs and handling
This is by a wide margin the heaviest lens in the selection. In fact, it is the only lens of this group to be heavy enough to lead to a front-heavy combination on a Sony ⍺7-type body. It is also the only lens among these which is known to be radioactive and has the tell-tale yellowing (which can however be treated). At f/2 it is not a blazingly fast lens, and due to the yellowing its actual transmission (t-stop) is the weakest of the lot. As a point of trivia, it is one of only a few lens designs (ever) to sport a clearly concave front element.

The lens’ build quality is fantastic. While the focus ring is somewhat heavy, its action is smooth and satisfying. The aperture ring sports a full range of half-stop clicks, and is decently tactile. Even though the lens is from 1971, it sports a bayonet attachment for a lens hood. The lens’ silver/chrome front (early FD lenses had those) makes it visually distinct.

Being an FD lens, the Canon can be finicky to use as one needs to remember to put (and keep) the adapter’s aperture engagement ring in “lock” to allow the lens’ aperture ring to have an effect on the lens’ diaphragm – a point of potential failure during a hectic shoot.

Image quality
Obviously, this copy with its yellowed thoriated element resulted in overall warm-tinted pictures which – while not displeasing – necessitate extra steps in post-production to address. Also, due to the yellowing, the lens’ actual transmission (t-stop) was worse than that of its brethren.

Otherwise, the lens’ optical performance was impressive. Besides the yellow tint, the FD’s optical performance was devoid on any clear weaknesses and especially its wide-open performance (especially at longer distances) was quite breathtaking.

Carl Zeiss Jena Prakticar 35 mm f/2.4 MC

Basic specs and handling
The Prakticar / Flektogon on the other hand is the third least heavy lens in the test and the second lightest legacy lens. As it simultaneously is a relatively small lens, it feels like a lens you would have few reasons to leave out of your bag. Designwise, the ‘Flek’ is the least bright of the entire test and has a relatively simple optical design: 6 lens elements in 6 groups and a 6-blade aperture. Given the significance of those numbers, maybe Ronnie was onto something.

Having handled a number of east-bloc lenses (including the MIR in this comparison), I must say, that this sample of the ‘Flek’ does not correspond to the phrase ‘built like a tank’ (often associated with east bloc lenses). In fact, compared with any major-brand 70’s Japanese lens, the Prakticar exudes a somewhat less rugged feel. With this sample, my main gripe is the feel of the aperture ring – especially the fact that the numbers and stops do not line up well enough (which I believe to be an inexcusable lapse in QC) – something which is blatantly obvious when used on a Praktica B/BC series body (which – alike many manual -age cameras has a small window looking down on the aperture ring, allowing you to see the selected aperture without removing your eye from the viewfinder). Otherwise the lens feels very nice and especially the focus ring is nice, smooth and easy to find in the dark.

Image quality
As I already alluded to, I included this lens in the comparison (but not a number of f/2.5 lenses), because it has a – relatively speaking – fearsome reputation: Whenever you ask around for a fast 35mm, many will proclaim their undying devotion to the Flektogon. In that sense, the Flek entered this comparison with high expectations resting on its narrow filter ring.

And while the Prakticar/Flektogon is optically not the best lens in the comparison, it does a really good job and performs very evenly. This is especially remarkable, because it performs well in especially those situations (CA’s, night-shots) when you would not expect an old 6-element design to perform. The Flek’s one real weakness is associated with its most obvious strength: while it has the shortest MFD (leadning to the highest magnification) of the lot, its corner sharpness at shorter focusing distances is somewhat frown-inducing (from f/2.4 to f/5.6).

Konica Hexanon AR 35 mm f/2

Basic specs and handling
Although at 286 g the Konica is of median weight, it is one of the larger (by volume) lenses in this test. Given that the authority on Hexanon lenses (buhla.de) gives it a weight of 320 grams (34 grams more than this sample), I suspect this last iteration has been significantly weight-trimmed, resulting in a lens which feels like a solid chunk of optics, without being especially heavy.

Personally, I very much like the Konica design philosophy, and this lens does not disappoint (even though I am partial towards the older Konica models with milled metal focus rings). This lens looks solid and has a nice form, the focus ring is wide and grippy, and the focus throw is more than decent. Sadly, this is one of the many Hexanon lenses which – while they sport an aperture ring – were not designed for on-lens aperture manipulation (Konica was one of the pioneers of shutter-priority auto-exposure. Hence, shooting aperture was designed to be set by the camera body). Not surprisingly therefore, the aperture ring is not a pleasure to use.

Image quality
The optical design of the Konica Hexanon 35/2 stayed the same (coatings were obviously brought up-to-date) throughout the entire history of the Konica AR mount (1971-1987). Moreover, optically this is the same lens as the Konica F-bayonet 35/2 introduced in the early 1960’s. In short, the lens was good enough to survive two photographic revolutions without needing a total redesign. And even in this comparison – where it is up against some very stiff competition – the Hexanon 35/2 makes a good showing.

In this comparison, the Hexanon has very good resolution throughout. At close range (0,5m) the results are outright brilliant and hold up better than most when the aperture is wide open. Alike many other lenses in this comparison, shooting wide open leads to a lessening of contrast, but the results remain more than usable. Stopped down to f/2.8, contrast is up there with the best of them. While the Hexanon did not cope as well in the night-time shots, showing signs of complex astigmatism, I do appreciate that the Hexanon managed to put in a very balanced performance throughout.

Minolta MD W.Rokkor 35 mm f/1.8

Basic specs and handling
The Minolta lens has a lot going for it. Firstly, it is one third of a stop brighter than f/2, being second in brightness (in this comparison) only to the Nikkor Ai 35/1.4. Secondly, it is a small lens… tiny even considering that it is brighter than all but one of the other lenses. Build quality is on par with what you’d expect from a Japanese major brand, with the obvious caveat, that many prefer the more solidly built MC lenses over the newer MD lenses. Mechanically, the lens works like the well-constructed piece of precision optics it is. Even so, there are two things I personally dislike about the Minolta 35/1.8. The first, and IMO unforgivable lapse, is the lack of any intermediate click-stops between wide open and f/2.8. Secondly, the tapering (narrowing) construction of the Minolta MD 35/1.8 leads to that I constantly fumble for the focusing ring, often (also) ending up with putting my fingers into the frame.

Image quality
Make no mistake. This is a good lens. Very good. It manages to produce a sharp, contrasty picture with good colors and decent amounts of microcontrast. Aberrations are also generally well controlled. Had I not sworn off the term “3D pop”, it might have found use here. Furthermore, the Minolta does not care whether you shoot daytime or nighttime, it delivers.

But even the Minolta has some weaknesses, and these become especially evident at wide open: Firstly, of all the lenses in this comparison, the Minolta clearly has the heavies wide open vignetting. Moreover, vignetting does not really go away before between f/4 and f/5.6. Secondly, and IMO more problematically, the Minolta suffers suffers severe contrast loss (worse than with the other lenses) wide open. Thirdly, at f/1.8 and close range (at 0,5m) the Minolta shows clear signs of a complex field curvature leading to that, oddly, sharpness and contrast are marginally worse in the image center than in the off-centre doughnut, before hitting rock bottom in the extreme corners. To add insult to injury, the extreme corners stay bad until about f/5.6, only to be among the best at smaller apertures than that.

Based on this sample, I’d say that if your prime intention for buying a fast moderate wide-angle is to use it mostly at larger apertures, or even wide open, then there are more balanced and less risky options. If, however, you can live with these limitations, the the Minolta is a nugget of 24K.

MIR 24N – 35 mm f/2

Basic specs and handling
Soviet lenses were mostly not known for brandishing low f-numbers, but the MIR is an exception. The mere concept of a Soviet-built Nikon F-compatible fast 35 mm lens is hugely attractive, especially as it is one of the most affordable propositions (among fast 35 mm lenses) out there.

Call me old-fashioned, but I like the MIR’s user interface. The lens is big, but not too big. Solid, but not too heavy. The control rings move as they should and are very tactile. I especially appreciate the existence of half-stop clicks all the way from f/2 to f/11 (whereafter full stop clicks to f/22). This is sensible. The 280 degree focus throw might feel excessive, but this lens also has a very short MFD. Moreover I like the “clean barrel” form of the lens and the somewhat ‘oldskool’ look and typography (admittedly, YMMV) .

Image quality
As you see above, I would have wanted to like this lens. Sadly, based on the sample I have, I cannot really recommend this lens. It’s not that it is really a bad lens, it’s just that it is also definitely not a good lens. Compared to it’s brethren, it is consistently weak. It is not as sharp and contrasty wide open, and it does not really sharpen up when closing down (even at f/11, it clearly trails behind). It shows significant barrel distortion (admittedly, the Porst is worse), and suffers from pronounced chromatic aberrations wide open. Moreover, they do not really go away even on closing down. Interestingly, once night falls, the MIR regains some competitiveness and is no longer the abject laggard.

Now, in case you’re wondering how, after this list of sins, the MIR is not a bad lens, the answer is quite simple: it is not a bad lens, but it is also not a good lens. Even with all these weaknesses this sample evidenced, the MIR is perfectly capable of taking very fine pictures. It’s simply, that they do not hold up to pixel peeping as well as pictures from the other lenses…

Nikkor Ai 35 mm f/1.4

Basic specs and handling
Obviously, this is an outlier lens. For older 35 mm focal length lenses, f/2.8 was pretty much par for the course, meaning that f/2 is at least a birdie and f/1.4 is an eagle, or maybe even an albatross. As such, this was never a ‘normal’ lens, and while today a 35/1.4 SLR lens is no longer the earth-shattering event it was in 1970 (when the predecessor version of this lens was launched), one must expect that to achieve a 35/1.4, Nikon engineers had to make some compromises (even though none are mentioned).

The most obvious compromises were in size, weight and price. While I do not have access to original price information for Nikkor lenses, I would be surprised if this had not – by a considerable margin – been the most expensive of these lenses (when new). Also, while all modern 35/1.4 lenses are both heavier and bulkier than this old Nikkor, it nevertheless is larger and heavier than all other lenses in this comparison (except the Canon FD).

On the other hand, the build quality is second to none. The focus ring is perfectly placed, decently wide and grippy and the aperture ring is a joy to use (except that – like all Nikkor’s – it has only full-stop clicks). Its shape is somewhat odd (the nose tapers from the wide focus ring to the 52 mm filter ring), exacerbated when you fit it with the intended HN-3 lens hood (a good investment considering that the front element almost protrudes beyond the filter ring).

Image quality
In some respects, this lens reminds me of the Leica tri-elmar lenses (unlike the vario-elmars, the tri-elmars are not zooms, but more like three primes in one), as I find this to be two lenses in one. Let me explain:

Firstly, thanks to implementing Nikkor’s close-range-correction (CRC), this lens shows quite different performance at close range and at (near to) infinity, as made obvious by the fact that this lens (alone among this comparison) has clearly different distortion characteristics at close/distant range. With an MFD of 0,3 meters, this lens is by no means a macro lens, but the CRC produces very good close-range performance.

Secondly, when shot wide open, this is an extreme lens. Based on this sample and the thousands of shots I’ve made with it, I must conclude that close range and long range this lens changes character and on the different sides of that divide, we’re dealing with two totally different animals.

At close range, f/1.4 is soft and glowing. There’s a lot of detail, and you can dig that up in post, but without some post-processing, the result is … risky. Don’t get me wrong, it has a character which can serve you very well, but calibrating that character to your benefit is not an easy task. If you don’t succeed, the result can at best be called ‘quirky’. However, once you move past f/2, you get a lens that produces sharp images with good contrast and immense detail, combined with a lack of field curvature which must probably thank the CRC-system.

At longer range, and smaller f-stops (f/5.6 ->) what you have is a lens which does show a wealth of detail and with good-to-great contrast, but does not always match its compatriots in raw resolution and is quite prone to CA’s in high-contrast transitions. On the other hand, at wide open (f/1.4) you have a lens which’s usefulness is determined only by how much softness you can bare, and between f/2 and f/4 you have a huge, expensive lens which is easily upstaged and outperformed by quite pedestrian rivals.

In sum, if you can spare the space in your lens collection/camera bag for a special-use lens (and can spare the time for learning its strengths), this lens is surely a respectable addition to your repertoire. If, on the other hand, you want to make do with one 35 mm prime, this is likely not that one lens.

Nikkor Ai 35 mm f/2

Basic specs and handling
Not only has Nikon had a 35mm f/2 lens in its lineup since 1965, very little about this lens changed until Nikon discontinued the manual version in 2005, 16 years after its successor (see below) had been introduced. Yes, changes were made during that production run, coatings were upgraded regularly, a rubber focus ring was introduced, and finally even the focus throw was shortened, but the essential – the optical design – was not touched. 8 elements, 6 groups, and 7 aperture blades was a winning recipe.

Not surprisingly, handling is second to none. Starting with the shape of the lens, everything is as close to perfect as Nikon could make it (I personally disagree with Nikon’s habit of eschewing half-stop clicks). This lens simply is a joy to use.

Image quality
So, how did it compare? Mostly, very well. While lacking the Canon’s wide-open corner performance, it’s miles ahead of the Minolta in the same respect. While lacking the Konica’s (and Canon’s) close range center performance at medium apertures it manages to resolve astonishing detail and produces very life-like imagery. If this sample is anything to go by, the Nikkor Ai 35/2 has only one weakness, and that is in extremely harsh contrasts: both in the night-time cityscape and the starscape the lens showed not only the sagittal astigmatism to be expected from a 1960’s design, but also showed a modicum of light scatter. Again, whether these are field-relevant issues is up to you to decide.

All in all, a very respectable result from one of the Japanese great lensmakers.

Nikkor AF-D 35 mm f/2

Basic specs and handling
In so many ways, this lens is all that which legacy lens aficionados want to get away from: besides a rubber focusing ring, the entire outer shell of the lens is plastic (including filter threads), the focus ring is barely dampened, and the aperture ring is not easy to grip, the click-stops are too pronounced, and the ring is relatively heavy to use. From a tactile point of view, this is not a nice lens.

On the other hand, with a design dating to the late 80’s, this lens is definitely not a current lens. Moreover, this lens is still an entirely manual contraption (in contrast to every 35 mm lens Nikon has introduced since). Yes, it has autofocus, but as it is the slot-drive kind and the lens itself does not contain a focusing motor. More importantly, while it has electronic contacts, the lens works perfectly well without them. This is especially relevant considering that the aperture works just as intended on any Nikon body introduced since 1977 (as well as any mirrorless camera using just a dumb adapter). It even has pilot holes in the aperture ring to facilitate installation of ‘rabbit ears’ to ensure compatibility with every Nikon body introduced since 1959! And thanks to the generous flange focal distance of the Nikon F-mount, you can use this lens on almost any digital interchangeable lens camera (given the right adapter).

Image quality
This lens has, from a purely optical standpoint, a mixed reputation. Reviews often hate it, pointing out vignetting and corner softness; users often adore it. Often, but not always. Some users (both old-timers and legacy lens fans) remember to point out not only its deficient usability (see above), but also that its optical design is comparably unambitious: with only 6 lens elements in 5 groups, this lens is far less sophisticated than its predecessor, the 8 element 6 group design used by Nikon from 1965–2005 in – among others – the Ai version summarised above.

Even so, Nikon clearly did something right in this simple design. Firstly, a combination of modern coatings and a low element count result in consistently fast shutter speeds indicating good transmission. Secondly, even though one could assume that the low element count would lead to below-par corner performance (especially wide open), this is not so. Granted, the AF-D is not at its best in corners, but neither is it bad. Not only does it produce decent results, but it is significantly better than several of the reviewed samples. Moreover, it tends to more than hold its own against its older (more sophisticated) sibling. Admittedly, the AF-D is not stellar before closing down to f/5.6, but from that point onward it produces imagery with definition and contrast which leave very little (if anything) to be desired. Moreover, if I were intent on low-light, high-contrast (nightscapes, starscapes) photography, and I wanted by result to be other than artistic, this is definitely the lens (of these) to choose, as its astigmatism is significantly more controlled.

Porst WW 35 mm f/1.8

Basic specs and handling
The Porst, with its 58 mm filter diameter (same as the MIR) is one of the thicker lenses in this sample. At the same time, it is relatively short (only the Nikkor AF-lens is shorter). Also, at 264 grams, it is not a particularly heavy lens. Even so, it feels solid enough, and ‘flimsy’ is not a word which comes to mind. That said, with its general “Japanese off-brand lens” -look it does lack that kind of ‘refinement’ you’ve learned to expect from a major brand-lens.

In sum, I find the Porst to handle very nicely, with two major caveats: Firstly, being an auto-aperture M42 lens it has that dastardly and error-prone A/M switch (which actually ruined one set of pictures even in this comparison). Secondly, it (like the Minolta) lacks any intermediate click-stops between f/1.8 and f/2.8.

Image quality
Before I go into the details, let me just remind you that this is a lens I’ve had to do considerable work on and sample variation may thus be an even bigger issue. Even so, the fact that the lens performed this well is – I think – testament to that I did not foul up badly during my rescue/repair.

With that out of the way, this is not a bad lens. Not at all. With a max. aperture of f/1.8, it is one of the brightest legacy 35 mm lenses available. Moreover, in these tests, the Porst also systematically recorded short shutter speeds, indicating that this brightness is not just paper brightness, but also translates as t-stops. More impressively, the Porst vignettes very little – wide open its vignetting is on par with the best of other tested lenses. At the same time, while many of these lenses produce really fuzzy and smeared extreme corners when used wide open, the Porst certainly loses some sharpness and contrast, but outperforms the majority comfortably (only the FD outperforms the Porst in extreme corners wide open.).

Nevertheless, the Porst has its weaknesses: Its distortion is higher than any of the other lenses, and while this distortion is easily corrected, and far from the kind of egregious distortion zoom-shooters are accustomed with, it certainly is on a level which often necessitates a few clicks in post. Also, the Porst lens evidences a degree of chromatic aberrations at all aperture settings: CA-wise, its sweet spot is at f/5.6, and going in either direction, it gets worse. While this CA easily shows up while pixel peeping, it is in some cases liable to show up in prints (beyond 30×20). Also, the lens’ longish MFD can be somewhat limiting.

Conclusion

Let me tell you some of what I’ve learned through doing this comparison. Firstly, there are more ways than one to make a good, fast 35mm lens.

Secondly, with one exception, each of these lenses have a specialty – something they excel at. This means, that whenever I go for a specific shoot, there is a lens or two which are especially suited to that usage scenario. Problematically, there is not a single lens among these which would be all-round splendid. Hence, unless your photography is very specialised, you will most likely want two or more fast 35 mm lenses.

As a corollary, while a lens may be excellent for a specific task, that does not mean that it will be of use to me, as I do not do all kinds of photography (and, I guess, neither will you). Therefore, while all these lenses are so good that they deserve to be used, I find it unlikely that I will hang on to all of them.

I hope you have enjoyed this comparison/review. If you have any comments, please leave them below, or drop me a line.

Back?

• Part 1: Introduction, the lenses: pedigree and handling
• Part 2: IQ-comparison I – The Brick wall test
• Part 3: IQ comparison II – Urban vistas
• Part 4: IQ-comparison III – Bokeh and blur
• Part 5: IQ-comparison IV – Night-time vistas
• Part 6: Summary and conclusions (you are here)