|
|
|
|
Navigation
|
Carl Zeiss Lens Performance Data Sheets
by Robert Bumb (rrbumb@nwlink.com) A Performance Data sheet is included with each new Carl Zeiss lens. Often times it is not immediately clear as to how the performance graphs should be interpreted. Explanations of the performance data can be complicated and highly mathematical (1, 2). Carl Zeiss stakes its reputation on the optical quality of its products. The photographer pays premium prices for Carl Zeiss lenses. The Performance Data Sheets, if you will, constitute the proof of the pudding with respect to optical performance. Not all lens manufacturers furnish such performance data for their lenses. Some furnish no such data at all. Some furnish data calculated from the design parameters of their lenses, but furnish no data from production examples of their lenses. Some furnish calculated data in such a way that it is impossible to determine whether the data take into account the natural loss of illumination. Zeiss follows the old manufacturing adage, “You get what you inspect, not what you expect.” Zeiss claims that to the best of its knowledge, it is the only supplier that furnishes graphs based on data taken from actual production examples of its lenses (3). What follows is a simplified discussion of the Lens Performance Data Sheet. References to a rigorous treatment of the underlying principles are given so that readers can pursue the details if they wish. Relative Illumination GraphThe Illumination graph is discussed first as it impacts the MTF graphs.
Graph from Carl Zeiss where K=2 is
f/2 and K=4 is f/4 This graph plots the ability of the lens when set to a specific aperture to illuminate uniformly the film plane. The vertical graph axis gives illumination levels from 1 to 0. At 1, the lens fully illuminates the film plane, and at 0 the lens does not illuminate the film plane at all. The horizontal axis lists the distances in millimeters from the center of the lens with 0 being the center of the lens. All lenses demonstrate a decrease in illumination from their center to their edges (4). Wide-angle lenses demonstrate a greater decrease in illumination at their edges than do normal focal length and telephoto lenses. This is due to the fact that light enters a wide-angle lens at oblique angles that are greater than those for other lenses and the light must travel further to reach the film plane (5). Lenses that have very large maximum apertures show a greater lost of illumination at their edges at their largest apertures than do those with smaller maximum apertures (6). The loss of illumination from the lens center to its edges complicates the ability of the photographer to maintain the contrast between bright and dark parts of an image at the edges of the film frame. In addition, one’s ability to measure contrast differences is dependent on the level of illumination. It is increasing difficult to detect differences in contrast as illumination decreases (7). This loss of illumination impacts the MTF graphs. MTF (modulation transfer function) GraphsAt first glance MTF graphs appear overly busy.
Graph from Carl Zeiss The vertical graph axis lists the retention of contrast seen in the object to be photographed that is delivered to the film plane by the lens. Perfect retention of contrast is 1. No retention of contrast is 0. No lens can achieve perfect contrast retention (8). The horizontal axis lists the distances from the center of the lens to its edges, as was the case in the Illumination graph. Generally, there are three sets of data plotted with each set of data containing two different lines. These data sets are used to graph the retention of contrast between a series of alternating black and white lines that are the focus target of the lens. Zeiss chooses to plot data for one set of alternating black and white lines that consists of 10 black and 10 white lines per millimeter. The second Zeiss plot is for a series of alternating black and white lines where there are 20 black and 20 white lines per millimeter. The third Zeiss plot is for a series of alternating black and white lines where there are 40 black and 40 white lines per millimeter. It becomes more difficult for a lens to maintain the contrast between the black and white lines as the number of such lines per millimeter increases (9). In addition, the natural loss of illumination makes it more difficult to measure the retention of contrast, as was noted above. Note that the lens is maintaining a constant resolution, but the retention of contrast is decreasing. (The concept of acutance is not discussed here, as the author is not qualified to discuss the internal mechanics and output signals of the human eye.) There are two plots for each set of data. One line is dashed. The other is solid. The solid line represents a series of alternating black and white lines that if fully drawn would appear as the spokes of a wheel radiating from the center of the lens. The dashed line represents a series of alternating black and white lines that if fully drawn would appear as a series of concentric circles radiating from the lens axis. It is a fact of physics that the concentric circle lines that would be generated by a wide-angle lens are more affected by loss of illumination than are the lines representing the spokes of a wheel (10). Consequently, the retention of contrast shown by the solid lines decreases faster than does the retention of contrast shown by the dashed lines when using wide-angle lenses. This is what you see in the Carl Zeiss MTF plots. This is not true for normal and telephoto lenses as also can be seen from the Carl Zeiss MTF plots. Note that the either lens type lens continues to resolve the specified number of lines, but the retention of contrast decreases differently between different focal length lenses. The ability of a lens to retain contrast between alternating black and white lines improves as the lens is set to a smaller aperture size. This can be seen in the Carl Zeiss MTF plots. (For the purposes of this discussion the affects of diffraction are ignored.) Distortion CurveThe Distortion Curve is clear and requires no further explanation.
Graph from Carl Zeiss Using The MTF Graphs To Compare LensesHaving spent the time to examine the MTF graph in some detail, it is an
unfortunate fact that MTF graphs from different sources for lenses made by
different manufacturers cannot be compared to one another. The reason why MTF data from different sources cannot be compared to one
another is that the MTF test conditions between manufacturers and testing
laboratories are not constant. The MTF tests are very sensitive to the
light used to test the lens, the focus setting of the lens, the internal
construction of the lens, the effective focal length of the lens, the spectral
transmission properties of the lens, and the test equipment used as well as the
operator of that equipment among other things (11). One Carl Zeiss competitor uses calculated MTF data. These data
approximate the optical performance of a lens, but they do not have the same
importance as the data taken from actual lens examples. Another Carl Zeiss
competitor uses different frequencies of alternating lines. Another tests
using only a single wavelength of light as opposed to a defined white light. The MTF graphs produced by Carl Zeiss for any Carl Zeiss lens can be compared directly to one for any other Carl Zeiss lens as all Carl Zeiss lenses are tested in the same way. The MTF graphs presented at Photodo (12) can be compared to one another. The Photodo data are very useful, as Photodo has been testing all lenses for eight years in the same way with the same equipment and the same equipment operator. Comparisons of the optical quality between lenses from different manufacturers can be made using the Photodo data. Some caution needs to be exercised in comparing lenses from different manufacturers using the Photodo MTF data. The optical performances of a lens with a weighted Photodo MTF of 4.5 and another with a Photodo weighted MTF of 4.0 are indistinguishable to the human eye (13). Said another way, weighted MTF differences of greater than 0.5 are significant in judging optical quality differences. It is also important to remember that Photodo tests a single lens. It is possible that the single lens tested may have assembly or adjustment problems although such problems should be visible at the lens center position for any one data set. When the difference between weighted Photodo MTF values is greater than 0.5, the differences in performance will be visible. In general, the larger the projection or print, the more visible will be the differences. There are a number of possible reasons for the optical performance differences including resolving power, internal lens flare, internal lens restrictions, precision of manufacture and assembly, quality and consistency of the materials of construction, and lens aberrations among others. The Acceptance of MTF DataThere are some, maybe even many, photographers who insist that a lens cannot
be tested in a laboratory. They have several reasons why they insist this
is the case. Unfortunately, this gives rise to comparisons such as,
“…From a There are some photographers who insist that the MTF tests incorrectly measure the performance of a particular lens manufacturer or a particular lens. They have several reasons why they insist this is the case. This is a long-standing issue with those who are most interested in the use of lenses with very large maximum apertures when the lenses are set to their largest apertures. Lord Kelvin, one of the founding fathers of thermodynamics, said, "When you can measure what you are speaking about and express it in numbers, you know something about it." We have come to rely more and more on the use Kelvin’s statement in almost every decision we make. When used with care, with recognition of the limits of detectable differences and variation of data sources, and when not used as a hammer, MTF data are a very useful tool for the photographer. MTF data can materially assist a photographer in selecting a lens based on the measured optical performance of a particular lens type, lens manufacturer, focal length, or aperture size. Such a tool is always a benefit.
|
|
|
Lens Performance Data
