NIKKOR - The Thousand and One Nights No.63
The AF Zoom-Nikkor 28-80mm f/3.3-5.6G
Tale 63 covers the AF Zoom-Nikkor 28-80mm f/3.3-5.6G, the standard zoom lens for autofocus (AF) cameras, and the lens that defined the "lens kit" sales method. The year is 2001, and a new way of selling film cameras is being defined. Nikon releases the high-performance Nikon U (Nikon F65/N65) at a reasonable price suited to the times. It was around this time that the "normal zoom set/kit" method of selling cameras that has become so commonplace was defined. As a result, the price of lenses sold as a set with camera bodies was assumed to much less than half the price of the same lens sold on its own. These lenses became known as "kit lenses". In this Tale, we will delve into the secrets of the AF Zoom-Nikkor 28-80mm f/3.3-5.6G, a lens that contributed to the establishment of the "normal lens kit" method of sales.
by Haruo Sato
I. The evolution of the "normal kit lens"
Camera bodies and lenses have long been sold together in sets. Originally, it was common for camera bodies to be sold with fixed focal length (prime) lenses, such as a 50mm f/2 or a 50mm f/1.4 lens. Naturally, camera bodies and lenses for systems other than interchangeable-lens camera systems could not be sold separately.
Therefore, it was natural for consumers to expect that cameras and lenses would be sold together. However, not only can lenses be separated from interchangeable-lens cameras, but users can choose from an assortment of lenses to create a variety of camera/lens combinations. One of the joys of purchasing a new interchangeable-lens camera has always been consideration of the lenses to be used with it. However, to aid consumers with their camera purchases, manufacturers created the "normal set" that combined a camera body with their recommended normal lens. The lenses selected by manufacturers to fulfill this role came to be known as "kit lenses". With the evolution to zoom lenses, "normal lens kits" became "normal zoom kits". It was around 1984 that we began to see the gradual switch from the sale of lens kits comprised of fixed focal length lenses to those that were made up of zoom lenses. Advances in design and manufacturing technologies contributed greatly to this switch. These advances enabled smaller, less expensive zoom lenses that offered better performance, all of which made them good kit lenses. The successful development of aspherical glass elements that could be mass-produced around 1991 contributed to even higher magnifications, support for wider angles of view, and even greater performance for normal zoom lenses. Kit zoom lenses continued to become less expensive, offer greater value, and cost less until the year 2000 when the price of a normal lens kit was very low.
As our advance into the digital age has begun, we find ourselves in 2017 with a wide variety of choices in normal lens kits, and even the revival of fixed focal length lens kits with which this sales method originated.
II. Development history and the designer
AF Zoom-Nikkor 28-80mm f/3.3-5.6G optics were designed by none other than myself. I took over where Mr. Kiyoshi Hayashi, who was the master of the concave-convex two-group zoom lens design, left off in the design and development of NIKKOR normal zoom lenses (kit lenses). This AF Zoom-Nikkor 28-80mm f/3.3-5.6G was only possible through the development of a number of other lenses. Let's take a look back at the relevant history.
While some may say that the first kit zoom lens was a 43-86mm zoom lens, I am of the opinion that the kit method of camera/lens sales was not truly established, by any manufacturer, until the AF age. In Nikon's case, this would probably be the Nikon F-501 (Nikon N2020) kit released in 1986, which included the AI AF Zoom-Nikkor 35-70mm f/3.3-4.5S designed by Mr. Hayashi. I followed Mr. Hayashi's path, developing a hybrid aspherical glass element, and designing and developing the AI AF Zoom-Nikkor 28-70mm f/3.5-4.5D, both in 1992. Then in 1993, with an even smaller, lighter, and less expensive lens as my goal, I developed the AI AF Zoom-Nikkor 35-80mm f/4-5.6D, designed with a six-element structure utilizing an aspherical lens element. I followed that with simultaneous development of the AI AF Zoom Nikkor 28-80mm f/3.5-5.6D and a new version of the AI AF Zoom Nikkor 35-80mm f/4-5.6D in 1995. In 1999, I oversaw the development of a new version of the AI AF Zoom Nikkor 28-80mm f/3.5-5.6D as development leader, and in March of 2001, the start of the 21st century was marked by the release of the AF Zoom Nikkor 28-80mm f/3.3-5.6G, which I designed and developed myself. The lens was constructed of six elements in six groups, and was, at that time, the world's smallest and lightest normal zoom lens. I began working on the design of this lens in February, 1999, and submitted my report on optical design in March, 1999. I was immediately given a development budget, and I presented the trial production plans I had created in June of 1999. Mass production trials began in August of 1999, ending an incredibly short development period for that time. The transition to mass production was made in October of 2000. The March 2001 release of the lens preceded that of the Nikon U (Nikon F65/N65) only slightly. This lens type with optics that had the most minimalistic structure in its class, six elements in six groups, and a concave-convex two-group zoom structure, was patented in Japan in 1992 and overseas in 1994.
III. Lens construction and characteristics
First, take a look at the cross-section of the AI Zoom-Nikkor 28-80mm f/3.3-5.6G (Fig. 1). Please forgive me if the following is quite technical. This zoom lens has the simplest structure and is representative of the Nikon tradition. It is a negative-positive (concave-convex), two-group zoom lens. The lens is constructed of the minimal number of lens groups needed to eliminate chromatic aberration—two. The negative front element is a concave lens with an aspherical surface and the second positive element is a convex lens. A positive rear group has a powerful triplet arrangement, convex-concave-convex. However, rather than a simple convex-concave-convex structure, dividing the first convex group into two elements preserves brightness and serves to move the principal point forward. Therefore, it is actually a four-element, convex-convex-concave-convex Ernostar structure. This Ernostar structure is the key. The Ernostar structure is optimal for negative-positive two-group zoom master lenses, and is the simplest solution to achieving a structure with the fewest number of lens elements. However, as anyone might, I wondered if we couldn't achieve a structure that used three elements rather than four. This would enable a structure with even fewer lens elements. Wouldn't you think, for example, that we could replace the Ernostar structure's front two convex lens elements with a single aspherical lens element that would compensate for spherical aberration and lower coma? In short, we wondered if we couldn't put a triplet structure on a diet. However, the front two convex lens elements in the Ernostar structure not only compensate for spherical aberration and coma, they also generate positive distortion and move the principal point forward toward the object (as with telephoto structures). It is very difficult to achieve both the bending that moves the principal point forward and well-balanced aberration compensation with a single aspherical convex lens element. If a single aspherical convex lens were used, the front group would end up paying the price. We would have to move not just the rear group’s, but also the front group's principal point backward, and the front group's structure and bending would also be restricted. So why do we have to move the front group's principal point backward and the rear group's principal point forward? That is to preserve the spacing that changes with zooming, and ensure the largest possible zoom ratio. Until this time, a minimum of seven lens elements was needed to achieve a 35mm format zoom lens with a zoom ratio of 1.5–2.5× and a maximum aperture of around f/4. This NIKKOR lens, however, tore down this seven-element wall, achieving a six-element structure that also managed to offer better performance than was possible with the seven-element structure.
IV. More lens elements are a "necessary evil"
As a young designer, a number of outstanding predecessors and mentors shared with me their knowledge of optical design. Among the tidbits passed along, I often remember hearing that "more lens elements are a necessary evil". I have spent my entire career designing lenses to be as simple as possible and to utilize the fewest number of lens elements possible. Why do you think this is? Depending upon how they are used, lenses are able to compensate for specific types of aberration. However, at the same time, they also generate other types of aberration. Therefore, the less experience a designer has, the more they may think that simply increasing the number of lens elements will improve the design, but they then discover that aberrations are not compensated as well as they expected and find themselves caught in a dilemma. While some may think that increasing the number of lens elements represents the evolution of lens design, reducing the number of elements can help to rid the design of that which is not necessary. Lenses have reflective surfaces. Increasing the number of lens elements by just one increases the number of reflective surfaces by two. The greater the number of lens elements in the structure, the more veiling glare there is. This results in imaging that is not as sharp or clear. An increase in the number of reflective surfaces is also linked to increases in ghost and flare. So what if we apply a good coating? The answer is simple. No matter how effective a coating is, it can never completely eliminate reflectance. By reducing the number of lens elements by just one, reflectance for two surfaces can be completely eliminated. Have you never used a lens, such as one with a six-elements-in-two-groups Dagor (Doppel-Anastigmat Görz) structure or a three-elements-in-three-groups triplet structure, that would seem to offer poor aberration performance, but actually rendered images better than did a seven-elements-in-six-groups Gauss structure? This is the result of differences in sharpness and clarity caused by differences in the number of lens elements. The numerical value used to express this is the "T-number" or"T-stop". MTF cannot be used to determine whether or not increasing the number of lens elements with this lens would be beneficial or harmful. That is why I often use cemented lens elements and always try to achieve designs that use the fewest number of lens elements possible. It is clear that if aberration characteristics are the same, a structure comprised of fewer lens elements will be better. Structures with fewer lens elements also have the happy benefits of being smaller overall, having a smaller diameter (filter attachment size, etc.), and costing less. This is why I say "more lens elements are a "necessary evil".
V. Optical performance
Now let's take a look at the aberration characteristics of the AF Zoom-Nikkor 28-80mm f/3.3-5.6G. We'll begin at the wide-angle 28mm position.
Let's start with spherical aberration. Almost full correction, just slightly under correction, was applied. At the time of its release, I think the amount of spherical aberration remaining was on the low end. I took particular care with regard to astigmatism and curvature of field. Astigmatic differences have been eliminated as much as possible in both the sagittal and meridional planes for correction throughout 85% of the image height (around 85% of the entire frame). In addition, measures to prevent the generation of as much sagittal coma flare as possible have also been employed. Therefore, point-image reproduction characteristics are good and blur characteristics (bokeh) are pleasing. However, the degree of distortion is relatively high at approximately -4.8% at infinity. This could be easily corrected with today's image-processing technologies. However, during the film age when this lens was released, distortion was this lens' weak point.
Now let's take a look at performance around the mid-range 50mm focal length. Spherical aberration is fully corrected. As for astigmatism and curvature of field, astigmatic differences have been eliminated as much as possible in both the sagittal and meridional planes for correction throughout 70% of the image height (around 70% of the entire frame). However, a slightly high degree of curvature of field remains. At this focal length, distortion is almost completely eliminated. The amount of coma is at its lowest, and differences in the shape of point-image formation depending upon focal length are minimal. Point-image reproduction characteristics are good due to very little sagittal coma flare. I think that those who are particular about sharpness will achieve the most satisfying image quality by shooting at around the 50mm focal length and stopping down the aperture a little.
Now, how does it perform at telephoto positions? Spherical aberration is under-corrected and indicated with curving in the opposite direction. Coma tends toward inner coma with a core surrounded by flare. Distortion is 1% or less. Compensation for chromatic aberration is at its best. Point images are affected by inner coma flare, but I think that optically speaking, three-dimensional rendering characteristics are quite good. The position at which focus was acquired exhibits slightly low contrast, but the image is well balanced with a core and pleasing blur characteristics (bokeh).
Though a bit of an exaggeration, I think it is safe to say that I tried to achieve consistent sharpness at wide-angle positions, and pleasing image-reproduction characteristics that take three-dimensional rendering characteristics into consideration, primarily for portraits captured at telephoto positions.
VI. NIKKOR charm in a low-priced lens
I think that the quality of our least expensive mass-produced kit lenses is extremely important. That is because kit lenses are very often the first lenses used by those new to Nikon and its products. Surely most those purchasing their first SLR purchase a reasonably priced camera and lens kit. Not only are the NIKKOR lenses included in these kits those users' first NIKKOR lenses, but they also represent the entire NIKKOR lineup for those users. You could say that to many, kit lenses = NIKKOR. Therefore, kit lenses must offer the same superior quality and performance consumers have come to expect of the NIKKOR name. That sometimes turns the tables a bit. However, even those of us who develop kit lenses must consider this both our mission and our responsibility, just as the designers of NIKKOR lenses through the ages have done.
VII. Actual performance and sample images
Next let's look at results achieved with actual images. Details regarding performance at various aperture settings are noted. Evaluations are subjective, and based on individual preferences. Please keep in mind that my opinions are for reference purposes when viewing sample images and reading the evaluations.
Maximum wide-angle 28mm position
f/3.3 maximum aperture
The center of the frame is relatively sharp with very little sense of flare. Sharpness remains fairly consistent closer to the edges, and both resolution and contrast are at practical levels. Resolution is not conspicuously high, but it is sufficient to achieve a pleasing image. Resolution drops only at the extreme edges, where flare is exhibited. Very little color bleed can be seen.
f/4 to f/5.6
Stopping down the aperture to f/4 increases both contrast and resolution from the center of the frame to the edges. At f/5.6, sharpness is increased to the extreme edges of the frame.
f/8 to f/11
Even more consistent rendering is achieved throughout the entire frame. Most notably, resolution increases for quite pleasing image quality. Of all aperture settings, the best image quality is achieved at f/8 to f/11. An aperture setting of f/11 is best for landscape photos.
f/16 to f/22
Even more consistent rendering is achieved throughout the entire frame, but there is a clear drop in resolution.
At f/22 to f/32 especially, the effects of diffraction are visible and resolution drops.
Mid-range 50mm position
f/4.5 maximum aperture
Resolution is consistent to the edges of the frame, and images exhibit a degree of sharpness sufficient for practical use. However, slight flare at the edges of the frame result in slightly softer rendering. Resolution is not conspicuously high, but it does achieve fairly decent image quality. Even at this focal length, there is little color bleed.
f/5.6 to f/8
Stopping down the aperture to f/5.6 reduces flare and increases contrast. Sharpness throughout the entire frame seems to jump a degree. Flare is nearly completely eliminated at f/8, and contrast is good to the edges of the frame. Of all aperture settings, the best image quality is achieved at f/8.
f/11 to f/16
Consistent rendering is achieved throughout the entire frame. Contrast is greatly increased. At f/16 the effects of diffraction are slightly visible and resolution drops a little.
f/22 to f/29
Rendering is consistent, but resolution drops. The effects of diffraction are visible and resolution suffers.
Maximum telephoto 80mm position
f/5.6 maximum aperture
Flare is a little soft from between the center and edges of the frame, but resolution is okay. Rendering is still decent for pleasing results. The extreme edges of the frame are not sufficiently sharp due to flare. While there is little focal point color bleed, slight color fringing can be seen in blur.
f/8 to f/11
At f/8 flare is eliminated and resolution seems better. Reproduction of tones is also good without too much contrast. At f/11 even more consistent rendering is achieved throughout the entire frame. Of all aperture settings, the best image quality is achieved at f/11. An aperture setting of f/11 is best for landscape photos. The maximum aperture of f/5.6 is probably best for portraits. It seems that rendering at the maximum aperture of f/5.6 is similar to that achieved with the high-quality Gauss-type lenses of the previous generation.
f/16 to f/32
Even more consistent rendering is achieved throughout the entire frame, but resolution decreases. At f/22 to f/32 especially, the effects of diffraction are visible and resolution drops.
If sharpness is the goal, the best results would likely be achieved at an aperture setting of f/8 to f/11 at all focal lengths (positions). For portraits, I think that I would shoot at a focal length close to 80mm and an aperture setting close to maximum aperture.
Now let's confirm these rendering characteristics with some sample photos.
So that you may judge the characteristics of this lens for yourself, compensation for lateral chromatic aberration and axial chromatic aberration has not been applied, and image sharpness has not been enhanced.
Samples 1, 2, and 3 were captured at the wide-angle 28mm position, the 50mm position, and the telephoto 80mm position, respectively. The low-contrast Neutral Picture Control was applied, and aperture was set to f/8, which would probably be the aperture most frequently used with this lens.
These images were captured at settings that would achieve low-contrast tones, but you can see that the results achieved at each of the three focal lengths show sufficient sharpness with no noticeable deterioration in image quality.
Sample 4 was captured at the 28mm position and f/3.3 maximum aperture. The image exhibits sufficient sharpness, clearly reproducing the girl's hair, and surrounding plants and flowers.
Sample 5 was captured at 50mm at the maximum aperture of f/4.5. The image is pleasantly clear with stable sharpness. However, blur characteristics are, unfortunately, a little harsh.
Sample 6 was captured at 70mm at the maximum aperture of f/5.3. Naturally, the image captured at this focal length is also pleasantly clear with stable sharpness. However, blur characteristics are still a little harsh at this position.
Sample 7 was captured at 80mm at the maximum aperture of f/5.6. This image is clear, but rendered just the tiniest bit softer than are images captured at other focal lengths (zoom positions). As a result, blur characteristics are that much softer. That makes 80mm the optimal focal length with this lens for portraits with which background blur characteristics are a priority.
Sample 8 was captured at the maximum telephoto 80mm position at the maximum aperture of f/3.5, and at the minimum focus distance of 0.35 m. I forgot to mention that one of the primary features of this lens was its short minimum focus distance. It had the shortest minimum focus distance of any kit lens on the market at the time it was released. As you can see in the sample image, this made macro photography possible with this lens. This can probably be said to have been the result of the six-element structure (two elements in the first group). It is a pleasing image that exhibits the necessary degree of sharpness in portions that are in-focus—namely the flower's stamen. I think you can see just how pleasing this image is with its gentle bokeh that exhibits no tendency toward double-line blur.
VIII. 100 years
Nikon celebrated the 100th anniversary of its establishment in July, 2017. This is our 100th year! I began working at Nikon in 1985, and have spent the last 32 years designing optics. That means that I have shared in about one third of Nikon's history. During this time, I have worked with many noted designers, experts in trial production, photographers, quality assurance demons (laughs), and by extension, so many other strange and wonderful people. I used to wish that I'd been born just a little earlier, but I don't anymore. I have lived and worked in a time of great innovation. Having joined Nikon when auto exposure became a reality, I have seen transition from manual focus to autofocus, from film to digital, and from photos to video. I now think that I was able to work during very important times. Having passed its first 100 years so successfully, Nikon is now looking forward to its next 100 years. How will Nikon be involved with optical devices, cameras, and the photo industry in the future, and how will we lead? There is still work to be done, and we must continue to do our best.
These days, I find myself most interested in the optimization of image formation characteristics. I have studied image formation characteristics in the three-dimensional domain of the optical systems that are a physical phenomenon, and reported my findings at Nikon and elsewhere. My lectures have been extremely well received by my peers. I am of the opinion that current evaluation methods for imaging optics are not sufficient to judge the performance of lenses for the video age sure to come. As we all know, the subjects of both photos and video are three dimensional. That means that image formation should also be evaluated in three dimensions. The performance of imaging optics must be evaluated based on their three-dimensional characteristics. As for optical design as well, I had thought that a time when we are able to completely control three-dimensional optical characteristics would naturally come. For example, the study of "psychological colors" has become popular.
However, the psychological evaluation that has been cultivated, involving many people, is still insufficient for the photo and imaging world. Even in the world of optics, there is very little research that goes beyond that of point formation. I think that there is a pressing need for further study in the area of defocus. I hope that in another 100 years, future study of images that are psychologically pleasing will have borne fruit. We are looking forward to the next 100 years and the future of the imaging world.