Leica Monochrom-II, part 1, May 10, 2015
Geoffrey Crawley had two simple rules for a report on cameras or lenses or on any other photographic equipment. The first rule states that after reading a report the prospective buyer of some piece of equipment would understand what to expect from this piece of equipment. The second rule states that after reading a report the photographer would know if the use of this piece of equipment would suit the photographer’s style or would improve his photographic prowess. He was quite careful not to fall into the trap of equipment competition. He hardly ever made comparisons between cameras and only occasionally made comparative tests of lenses between several marques. The only exception were his extended reports about chemicals, films and papers in which he was quite careful to present the characteristics of the material without letting his personal opinions dominate the assessment.
The new Leica Monochrom, let us refer to this camera as the Monochrom II to distinguish it from the original Monochrom I, does deserve a dispassionate report because it is, as so many Leica cameras in the past, not a mainstream camera. The important question to answer is the topic of Crawley’s second rule: does using the Monochrom II improve my photography or how does my photography change when using this camera.
The main characteristic of the Monochrom (I and II) is the absence of the Bayer CFA filter. There are of course technical issues to discuss, but above all there is a way-of-life-type of discussion. One of the main characteristics of the digital camera equipped with a sensor covered with the color filter array (CFA) is its versatility during the picture taking process and during the post processing stages of the photographic workflow.
The sensor of the camera can only record light intensity, in the same way as the silver-halide grains do. To add color information to the recorded image, a CFA with Bayer layout is introduced. The image file is then composed of an array of pixel locations with two values: a color (RGB) characteristic and a numerical value representing the light intensity. Because every pixel is individually addressable, and the numerical values can be algebraically manipulated, there exists an infinite space of possibilities. Basically every pixel in the sensor array represents one single object point in object space. The reconstruction of the original color value of every pixel is the task of the demosaicing algorithm and there are many ways to accomplish this goal. The primary goal is a faithful reproduction of the colors of the object. We all know that a really faithful record is impossible, unless in a scientific environment. And the human perception adds its part to make color reproduction and interpretation a rather subjective matter. The fine adjustments possible with the numerical values, recorded in the image file, allow for an infinite variation of color hue and brightness. There is no rule for correctness here and every person has to choose what color space is emotionally satisfying and/or accurately true-to-nature. The colors are made up of a weighted sum of the luminance values of the three color components. When these values are made identical, we get a neutral tone, also known as a monochrome tone. There are a number of dedicated programs, like Exposure 7, Darktable, Silver Efex Pro, Accu Raw Monochrome and DxO Filmpack 5, that try to reproduce as accurately as possible the many flavors of color, slide and black-and-white emulsions that once existed or still exist. The CFA-filtered sensor array can be the base for a full range of ‘film-looks’. Additionally the possibility of the selection of a wide range of ISO speeds adds to the flexibility of the digital sensor technology.
The closest equivalent was the Kodak Ektapress Multispeed that could be used at ISO speeds from 100 to 1000/1600 and could also produce black and white images when suitable paper was used in the darkroom. This flexibility was lost when the photographer had opted for one of the several black-and-white film emulsions. There was no color option and the spectral sensitivity of the emulsion predestined the brightness values of the colored objects. The use of filters could change the tonal values within some limits. And the printing process in the darkroom could also change the tonal impression by the selection of the chemicals. The black-and-white photographer made a conscious choice for less flexibility, a limited tonal scale and a reduced level of influence in the full reproduction cycle. This reduction as well as this limitation forces the photographer to maximize his knowledge of the process and to approach the object with a clear idea about the final image to create or record. Detached from the (important) topics of definition, spectral sensitivity, tonal range/reproduction and dynamic range (contrast or characteristic curve), the choice for the Leica Monochrom II is primarily a choice for a narrow bandwidth of tools for recording the reality. Many painters have produced works with the technique of etching and engraving, because this technique with fine needles allowed the artist to concentrate on details and strong chiaroscuro effects. As a metaphor we might say that choosing the Monochrom II (or a black-and-white emulsion for the M-A)is like an artist who trades in his plate with brushes and paints for a set of needles and a simple metal plate.
The corollary of this choice is the acceptance of the way the camera maps a specific color hue and brightness level to a specific shade of grey. The photographer who now uses silver-halide B/W materials is confronted with the same situation. Every film has its specific spectral sensitivity profile: the same red in nature may be reproduced as a more or less brighter grey value when recorded by Tri-X or Delta100. A red filter in front of the lens will brighten the red parts of the scene, but will darken the green parts. The choice of a red or orange filter must be done with care, because such a filter is an effect filter and not a correction filter like the yellow one. The spectral sensitivity curve of the Monochrom II is not yet published and one needs to experiment to discover the effects one likes.
The gamut of computer-based manipulations is rather limited because of the absence of the RGB pixel information in the image file. The photographer who intents to use the Monochrom II should be aware that the balance of influencing the picture shifts from computer to the camera as in the days of the silver-halide processing.
The effects possible with the computer programs are restricted to the classical darkroom techniques of change of contrast, of exposure and cool/warm-tone choices. The Monochrom II is eminently suitable for the photographer who wants to spend as less time as possible behind the computer screen and wants to shorten the workflow from shot to (inkjet) print. One might even state that the Monochrom II forces (or less intimidating: invites) the photographer to shorten the workflow and to optimize the relevant parameters during the shooting session. The LiveView option might be seen as the classical Polaroid control picture before the shooting.
The basic sensitivity of the sensor is ISO320. This speed is rather high and when one wishes to use the high-speed Leica lenses wide open the choice of ND filters is often necessary.
To approach the Monochrom II as the digital implementation of the prime analogue process and method, implies to look at the definition first.
Above: MM-II, ISO 320
I still use the test charts that were excellently suited for silver-halide negatives: this chart gives detailed information about the performance of a lens/film combination. Because this is a test of the camera and its sensor, I have selected the best lens and its best aperture (Apo-Summicron-M 2/50 mm ASPH at f/5.6) at a distance of two and three meters to get the best possible results. I made several test runs on different days to ensure the result could be reproduced. The picture below is a small part of the test chart, engaged to 200% to present the results clearly. The distance from camera to chart was 3 meters and the figures have to be multiplied by 60 for purposes of comparison. At ISO 320/400 the MM-II delivers a fascinating 75 - 96 linepairs/mm. Note that the vertical direction is less well resolved than the horizontal direction. These values represent the limiting resolution and when the requirement of crisp edge sharpness is added the useable definition is still in the region of 60 to 70 lp/mm. This result has been achieved when the camera is on tripod and with the LV option and sharpness peaking info. By the way the rangefinder gives exactly the same results. The performance when handholding the camera at a shutter speed of more than 1/1000 gives almost identical results.
This performance stays on the same level even when selecting an ISO value of 6400. In this case the limiting resolution of 75 - 96 lp/mm still holds, but the edge contrast has been reduced. It is safe to state that the MM-II paired with the Apo-Summicron-M 2/50 mm ASPH is a formidable performer.
The obvious comparison is with the previous model, the Monochrom I (MM-I), and the current M with Bayer CFA (M). As already reported in a previous article, the MM-I has for all practical goals the same performance as the M. The Siemens star is well suited for comparisons because the converging radial spokes show clearly the differences in resolution and the aliasing effects. The picture made with the M shows that the color demosaicing has problems with the very fine detail that has to be resolved. The MM-I has almost the same resolution and shows a cleaner image, but the lower number of pixels limits the level of detail and the ultimate size of the picture when printing. The MM-I has a distinct performance jump when compared to the M9 and predictions, including my own, gave the MM-II the same comparative advantage. To my surprise this prediction did not materialize. When you look very closely at the images (here at 200%) there is a subtle improvement in the separation of the black-white spokes close to the centre, but this distinction, while visible in these comparison pictures, might be lost in practical photography. This is a slightly surprising result and one that needs more study and analysis to explain.
The next part discusses the tonal range and the noise pattern of the MM-II, again compared to the MM-I and the M.