Zenitar 0.95/50 (Sony FE) ultra-fast lens review and comparison test + computational bonus

Zenitar 0.95 / 50 lens material especially for Radozhiva prepared Rodion Eshmakov.

Zenitar 0.95/50 (hereinafter referred to as Zenitar 50/0.95) is a modern lens for full-frame mirrorless cameras produced in a small series at KMZ (Krasnogorsk, Russia). Lens development was started by an optician Vladimir Bogdankov back in 2015, when the full-frame Mitakon Speedmaster 50 / 0.95 had just appeared, and the first prototypes were already assembled in 2016, but, largely due to effective KMZ managers, the serial lens went on sale only in 2019, and only for Sony cameras.

Zenitar 50 / 0.95 by its very existence raises many questions related to the veracity of the declared parameters (the Chinese, for example, very often lie when specifying actual relative aperture), optical quality and artistic properties. At the same time, representatives of KMZ refused to provide their equipment for writing reviews, because Zenitar 50 / 0.95 avoided my attention for a long time. As a result, this lens (released in 2019) was provided specifically for the preparation of the article by the commission photo shop Kutuzov-Photo (Moscow).

Technical specifications

Optical scheme - 9 lenses in 8 groups, development of the "double Gauss" scheme ("Planar");

Focal length - 52 mm (accurate);

Relative aperture - 1: 0.95;

Field of view - 44 °;

Estimated frame size - 36×24 mm (full-frame lens);

Aperture - 14 blades, stepless, with manual control;

Aperture limits - 1: 0.95-1: 16;

Focus type - manual;

The minimum focusing distance is 0.7 m;

Thread for filters - 72 mm;

Camera mount - Sony E;

Weight - 1.1 kg.

Design and execution of the lens

The first thing you notice when you see the Zenitar 50/0.95 is its size. Among all full-frame 50 / 0.95 lenses for mirrorless cameras, it is the largest. The reason for this is simple - the lens uses a relatively simple nine-lens optical design (versus 10 lenses for Mitakon 50 / 0.95 or 11 lenses for ttArtisan 50 / 0.95 Leica M) without the use of aspherical components (2 for Leica Noctilux 50 / 0.95, 1 for ttArtisan 50 / 0.95 ), high-refractive (with a refractive index of more than 1.8, 1 for Mitakon 50/0.95, 8 for ttArtisan 50/0.95 Leica M) and low dispersion (4 for Mitakon 50/0.95, 2 for ttArtisan 50/0.95) glasses, so the dimensions were brought into sacrifice in the course of the calculation, since when correcting distortions in the case of a limited number of parameters, interlens distances play a significant role. It turns out that the Zenitar 50 / 0.95 is the simplest modern full-frame 50 / 0.95 lens in terms of optical design.

The exterior finish of the lens is pretty standard and discreet - the lens does not guess anything special or premium. The all-metal body is covered with a simple black matte anodization, the quality of which, however, there are no complaints. There is a scale of distances in feet and meters, a depth of field scale (I wonder what pixel size it is designed for?). In my opinion, the lens lacks some zest in design. On the other hand, it's good that KMZ did not copy Leica or Voigtlander or, moreover, invent "elegant" solutions like the bayonet "skirt" of Zenitar 35/2 and 50/1.5 lenses.

On the front of the wide title ring of the lens, you can see the manufacturer's emblem, the name of the lens, parameters and an indication of the diameter of the filter thread. The latter is quite pleasant, because sometimes it is not immediately clear which filters fit the lens if there is no such marking, and there is nothing to measure the diameter with. Such a striking difference between the outer diameter of the lens and the size of the front lens is due to several factors. Firstly, the front lens of the objective is plano-concave negative (yes, the front surface is absolutely flat) and therefore smaller in diameter than subsequent ones.

Secondly, there is a retractable hood on the nose of the lens. To be honest, it would be better if she didn’t exist - she constantly hangs out, is not fixed at all in any of the extreme positions, and the benefits from her, as it turned out during the shooting, are homeopathic. Probably, the hood was added more for design reasons, in order to hide the difference in case diameter.

Of the other moving parts on the lens barrel, there are also useful ones - a wide focus ring and an aperture ring. The controls are ribbed for a comfortable grip, but there are no rubber pads on them. The lens focuses to a minimum distance of 0.7 m - this is the same as that of 7artisans 50 / 1.1 or Primoplan 58 / 1.9. In my opinion, such an MDF for a 50 mm lens is too large. On the other hand, the motivation for limiting the helicoid travel is obvious - reducing the weight and dimensions of an already heavy lens. In addition, it is unlikely that a 50 / 0.95 class optic will be used for Close-up / Macro shooting - much more modest lenses.

Aperture control is carried out steplessly, and you can see that the aperture setting scale has an unusual look - it does not “shrink” from F / 0.95 to F / 16, and the distance between the marks changes generally non-monotonically. The reason for this is the unusual Γ-shape of the aperture blades, similar to the aperture of lenses. Jupiter 6 и Helios-40. Apparently, KMZ decided that this was their "trick" and made the same for Zenitar 50 / 0.95, only with a large number of petals (14 pieces), due to which the hole is more or less round.

The Zenitar 50 / 0.95 aperture looks beautiful. However, its petals, although matte, are certainly not black, which hardly contributes to an increase in contrast on covered apertures.

The lens optics are covered with multilayer coating, but, knowing the real production capabilities, you should not count on more than 3 layers. However, there is nothing with an exorbitant refractive index in this lens, so even 3 layers would be quite enough to ensure good light transmission (but not perfect color reproduction).

Unfortunately, the lens is only available in one version - for E-mount. I prefer to use optics made for Leica M mount, since such lenses can be mounted on any modern mirrorless camera, and there are, for example, autofocus adapters Leica M - Sony E or adapters with a helicoid to reduce MDF. However, it would probably not have been possible to fit the Zenitar 50 / 0.95 rear lens into the Leica M mount, and there were no other promising full-frame mirrorless systems at the time of the start of development, and as a result, the factory did not master the production of Canon RF and Nikon Z mounts after the start of lens production. Let this KMZ not scare you too much - they didn’t master the DSLRs on LZOS and mounts, as a result of which everything "Rubinars" only available with M42 mount.

Tactile and visual Zenitar 50 / 0.95 did not leave any special impression. The lens is externally made concisely and without any gross flaws, with the exception of an unsuccessful lens hood.

Optical properties

A detailed consideration of the optical properties of the Zenitar 50/0.95 lens is of great interest, since, as noted above, this is the simplest of the existing full-frame 50/0.95 lenses. To study the properties of the lens (among other things), comparative testing was carried out with SMC Pentax 50 / 1.2 lenses and 7artisans 50 / 1.1 on a Sony A7s camera, as well as modeling the lens (based on the data of patent RU 2726264 C1) in the ANSYS Zemax software package.

Zenitar 50 / 0.95 and the problem of vignetting

When shooting with the Zenitar 50/0.95, the extremely strong vignetting immediately attracts attention - many users have already pointed out this drawback of the lens. It is especially bad that the vignette does not go away completely even when stopping down, which can be clearly seen when modeling the lens in Zemax - this means that the rear lens has an insufficient light diameter.

However, in reality, at F / 8-F / 16, the Zenitar 50 / 0.95 has not just dark, but literally black corners, as if the lens does not cover the frame at all, which is contrary to the simulation results.

When I randomly saw a photo of the lens from the side of the rear lens, I noticed that behind the rear lens of the lens there is a frame - a frame limiter, which is visually significantly smaller than the 36 × 24 mm frame and the rear lens itself.

I considered this frame to be the reason for the presence of excessive vignetting on the lens, but in order to verify this fact, I had to take the lens (this was really the main motivation) for a test and remove the frame limiter.

A simple test on a white wall at F / 16 confirmed the correctness of my guess - it is the frame limiter, incorrectly calculated by the KMZ engineers, that causes excessive lens vignetting.

Removing the frame doesn't completely solve the problem (and shouldn't, as the simulations showed), but the annoying opaque vignette goes away. Removing the frame limiter is easy - just unscrew the two slotted screws that fix the frame and pull it towards you. The solution may be useful for those using this lens. I took all the pictures with Zenitar 50/0.95 and all the tests with its participation with the frame limiter removed.

Testing the optical quality of the Zenitar 50/0.95 lens compared to SMC Pentax 50/1.2 and 7artisans 50/1.1

A simple evaluation of the sharpness of lenses in the far field was performed by shooting the same scene on camera Sony A7s (12 MP full frame matrix) and each of the lenses, followed by a review of 100% crop.

Below are examples of photos on 7artisans 50 / 1.1 with apertures from F / 1.1 to F / 8.

Next - a photo on Zenitar 50 / 0.95 (F / 0.95-F / 8).

Finally - a series of photos on Pentax 50 / 1.2.

Let's look at 100% cropped images taken at the widest aperture - F/0.95 for Zenitar, F/1.1 for 7artisans and F/1.2 for Pentax.

It can be noted that in the central part of the frame, the 7artisans 50 / 1.1 lens has the best image quality - it is distinguished by high contour sharpness compared to other lenses, and fairly good detail reproduction. Worst of all in the center of the frame is the old classic Pentax 50 / 1.2, which suffers from strong spherochromatic distortions.

At the same time, 7artisans turned out to be the worst on the edge of the frame, but the image obtained from Zenitar, although it looks smoky and hazy, retains much more details than the Pentax 50 / 1.2 image. Nevertheless, visually the Pentax 50/1.2 picture is perceived better along the edge.

Thus, in terms of the balance of sharpness at an open aperture between the center and the edge of the frame, the Zenitar 50 / 0.95 looks better than its rivals, while it is also the fastest of the test subjects. Moreover, the notorious vignetting of Zenitar in the absence of a frame turned out to be lower than that of 7artisans 50 / 1.1.

Further, the 7artisans 50 / 1.1 and Zenitar 50 / 0.95 lenses were apertured to F / 1.4, and the Pentax was left unchanged, since there are no marks for setting F / 1.4 on its ring.

With aperture up to F / 1.4, the sharpness of both modern lenses in the center of the frame was equal on my camera, and Pentax in the center is noticeably inferior to them. We also note that at the edge of the frame, Zenitar at F / 1.4 overtook Pentax at F / 1.2.

When lenses are apertured to F/2, the old Pentax is still noticeably inferior to 7artisans 50/1.1 and Zenitar 50/0.95. Most likely, the picture of new lenses is characterized by a small contribution of third-order aberrations, but is determined by higher-order distortions, which are more dependent on aperture and therefore disappear faster with aperture.

 

At f/2, the Zenitar clearly outperforms its rivals in terms of image quality at the edge of the frame.

With an aperture of F / 2.8, I no longer see any differences between the lenses in the central area of ​​\uXNUMXb\uXNUMXbthe frame on my camera, so I did not consider this part further.

As for sharpness at the edge of the frame, the Pentax 50 / 1.2 managed to catch up with the Zenitar 50 / 0.95 at F / 2.8: the resolution of the Zenitar limits lateral chromatism.

Exactly the same situation is observed at F / 4 - and the old Pentax is already starting to overtake Zenitar in sharpness at the edge of the frame.

At f / 8 aperture, Pentax has the best image quality across the field. Zenitar inferior due to the presence of lateral chromatism. And 7artisans 50/1.1 is just starting to bounce back, but it is still very far from the ideal, which it will never reach.

Thus, the Zenitar 50 / 0.95 has an optical quality much better than the old aperture fifty fifties like 50 / 1.2 or 50 / 1.4, in the aperture range up to F / 2.8. Also, the Zenitar turned out to be a much better lens in the field than the modern 7artisans 50 / 1.1. However, the Zenitar could hardly stand up to comparison with a lens like the ttArtisan 50/0.95 (Leica M), which has a much more complex optical design.

In general, it can be noted that the Zenitar is characterized by a good reproduction of fine details even at an open aperture at the edge of the frame, but at the same time, contour sharpness in the center at F / 0.95 is rather compromise, and at the edge of the frame it is simply absent. By F/1.4 the lens is quite good within the APS-C area of ​​the frame, and at F/2.8-F/4 it reaches its maximum performance.

It is also worth noting the very low level of distortion compared to 7artisans 50/1.1 (noticeable "cushion") and Pentax 50/1.2 (noticeable "barrel") and a very good correction of longitudinal chromatic aberrations - when shooting, it really rarely happens that you have to correct the color fringe . But, apparently, for a good correction of the longitudinal HA I had to pay with a high level of transverse, spoiling the picture on covered apertures.

Comparison of artistic properties of lenses Zenitar 50/0.95, 7artisans 50/1.1 and SMC Pentax 50/1.2

In order to study the features of the Zenitar 50 / 0.95 picture, I conducted a small photo session using three lenses indicated above at the same time, during which each scene was shot with an open aperture, a fixed white balance and manual mode. As a result, it was also found that when using equal values excerpts with an open aperture, the photo on Zenitar 50 / 0.95 turns out to be 1 / 3-1 / 2 steps exposure lighter than the photo on SMC Pentax 50/1.2 in the center of the frame. Thus, aperture Zenitar 50 / 0.95, most likely corresponds to the declared.

Below is a series of shots on 7artisans 50 / 1.1 on an open aperture.

Next - a series of photos on Zenitar 50 / 0.95.

And photos on Pentax 50/1.2. The missing shots turned out to be rejected in focus - the lens has a too small ring in diameter with a small angular movement for such an aperture, so the focus is easily lost and I had unexpected errors when shooting with this lens.

You can see that the shape of the disks in the bokeh of the lenses differs markedly from each other. In the central part of the frame, on the optical axis, at the Zenitar, the focal disk is characterized by the absence of a pronounced border, but the asymmetric border oriented from the center of the frame becomes more pronounced the further the disk is from the axis. In this case, the shape of the disk becomes domed, and the size of the spot sharply and rapidly decreases from the center to the edge, which is associated with strong vignetting and the absence of a pronounced field curvature. In my opinion, this behavior of bokeh is undesirable - a 50 / 0.95 lens is largely taken for the sake of blur - strong background blur, interesting background blur. It seemed to me that the Zenitar 50/0.95 visually does not differ from a good 50/1.4 lens in terms of the degree of object separation from the background, and does not differ from a bad 50/1.4 lens in terms of blurring.

Pentax 50/1.2 blur is marred by pronounced spherochromatism, which causes the bokeh discs to have a bright blue-green border and a frankly red middle. In the corners of the frame, the spots turn into triangles, a more or less pleasant blur can be obtained at distances of a half-length portrait, no more.

The complex and bright bokeh of the 7artisans 50/1.1 is very reminiscent of the blurring of the background of the classic ancient Zonnar of the 30s - this is both its plus and its minus. One way or another, it is strikingly different from the bokeh of both planar rivals.

User experience

When I took the Zenitar 50/0.95 for shooting, I expected it to behave approximately the same as the 7artisans 50/1.1, only worse - namely, strong chromatism, poor sharpness at the edge of the frame. At that time, by the way, I had not yet modeled it in Zemax and did not know for sure how it was fixed.

What really surprised me was the ability to shoot wide open without a clear central composition, which is required, for example, when shooting on 7artisans 50 / 1.1 or some Helios in most cases due to a poorly corrected field. In other words, Zenitar is good enough on the edge of the frame even at an open aperture. A pleasant result is obtained when shooting at an aperture of F / 1.2-F / 1.4, when contour sharpness returns to normal.

In general, I liked how quickly the lens picks up sharpness with aperture - not every ultra-fast fifty dollars gives acceptable quality across the field even at f / 4. Yes, the very corners of Zenitar never return to normal, but there are no complaints about the edges of the frame. You can also note the lack of focus shift - that is, the displacement of the focal plane during aperture. In fact, this, one might say, is a feature of lenses designed by Vladimir Bogdankov, who considers focus shift to be an unacceptable defect for photographic lenses.

The lens pattern, it seems to me, is with character, but “for an amateur”. Bokeh is generally a subjective thing, but still it was worth working out in the calculation on purpose, since “bokeh” and “50 / 0.95” are things with a more obvious connection than “sharpness” and “50 / 0.95”. I would say that Zenitar 50 / 0.95 is a good technical super-aperture lens, but hardly an artistic one.

More sample photos on a full-frame mirrorless camera Sony A7s are presented below.

Computational Bonus: Calculating the Ultrafast Zonnar and Comparing It to Zenitar 50/0.95 in ANSYS Zemax

Despite the presence of 7artisans 50 / 1.1, the idea of ​​\u50b\u1.0bcreating a super-aperture Zonnar type XNUMX / XNUMX has not left me for a long time, and earlier I showed especially fast nine-lens lenses from night vision devices PNV-57E  и TVNE-4B with parameters 37/1.0 and 50/1.2, respectively. However, both of these lenses have too short a back-segment and do not work well with full-frame cameras. Nevertheless, their schematic diagram with a meniscus corrector was taken by me as a sample for calculation.

The starting point was the 50 / 1.5 lens of the calculation of Ludwig Bertele (the famous German optician, author of the Sonnar scheme) from US patent 186621 in 1938.

In addition to the insufficient relative aperture, this lens also has triple gluing in scheme 2, which is considered extremely expensive, complex, and in the case of the front element, an unnecessary element in the presence of multilayer coating. Despite the ugly appearance of spots, the lens, surprisingly, has a good contrast transfer function.

This lens from the patent was also important as a reference point for designing a picture of a future super-aperture monster - I tried to further reproduce the characteristic look of bokeh, balancing the aberrations literally by hand.

The limitations in the calculation were the number of lenses (no more than 9), material (LZOS glass), rear segment (for mirrorless cameras), rear lens size (for E, Z, RF mounts), frame coverage (36 × 24, without vignettes at a closed aperture).

I should note that the task of calculating such a lens is very difficult. I've spent days trying to get a result that probably doesn't even reach the level of 7artisans 50/1.1. As a result, I got a lens with parameters 55 / 1.0, very reminiscent of the original Sonnar 50 / 1.5 in terms of drawing.

In fact, there is a chance that this lens does not lose much to 7artisans 50 / 1.1, according to the contrast transfer function. First, the Chinese often show it in Chinese, then there are lies. Tritely, it can be calculated for a more “simple” spectral function. Secondly, the difference between F / 1.0 and F / 1.1 is still very large when performing the calculation. As between F / 0.95 and F / 1.0. Finally, unlike the 7artisans 50/1.1, this lens has more classic bokeh and less vignetting.

Moreover, of course, I am almost sure that the calculation is not optimal and the lens can be calculated in the best way. Thus, the main difficulties were caused by the search for a compromise between the correction of chromatic aberrations and field curvature: the former requires glasses with a relatively low dispersion (ve > 50), and the latter requires glasses with a high refractive index (ne > 1.75). The problem is that there are no such glasses in the LZOS glass catalog, which leads to the need to use heavy flints with a non-optimal dispersion course and a too high refractive index and limits the corrective possibilities. In addition, there are difficulties with controlling higher-order aberrations generated by a lens with a large curvature at the rear of the lens - it is these distortions that, paradoxically, determine the optical quality of the Sonnar lens and its characteristic pattern.

Finally, I will also give diagrams for the Zenitar 50 / 0.95 lens.

It is easy to see how much better the Zenitar 50/0.95 is in image quality compared to both Sonnars - this is mainly due to the excellent control of wide beam aberrations (i.e. those that affect the image only at an open aperture) and excellent correction of longitudinal chromatism and curvature of the field. And it is also clearly visible that the Zenitar really “disappears” bokeh from the center to the edge of the frame - in many respects also due to the well-corrected curvature of the field. These observations not only stop, but spur the desire to understand better and still bring Zonnar's calculation to a really acceptable quality.

Conclusions

The most important thing is that Zenitar 50/0.95 is really F/0.95 unlike some Chinese products. The second is the simplest modern 50 / 0.95 lens in design. This is no longer the ancient 50 / 1.4-1.2, but definitely not Noctilux yet. Third, the Zenitar 50/0.95 has a sufficient level of optical quality to be usable. At the same time, of course, the picture formed by Zenitar is quite specific, but it will find its lover.

You will find more reviews from readers of Radozhiva here.