The always enthusiastic rumor sites now claim that the forthcoming Canon Professional dSLR will have a sensor with 54 Mp (by the way: a week ago they said that there would be no Canon professional DSLR introduction in 2011, now they claim there will be two!). Interesting is that the accompanying text claims that there is no low pass filter and that all L lenses are capable of using this ultra-fine pixel grid. Exactly the two aspects I discussed in the previous post. To do some calculations: 54 Mp amounts to a pixel dimension of 4 micrometers or 250 lines per mm or 125 lp per mm. I have great respect for the image quality of the Canon L lenses, but such a claim ( a useable resolution of 125 lp per mm with good contrast >20% even at the optimum aperture) is a bit over the top. It begins to look like the American big V8 engines in the 1950s cars that claimed hundreds of horsepower that was not available but could safely be stated as nobody on the street could test it or feel it. But is was nice to boast about with your car-owning-neighbor or your prospective girl-next-door.
That said I am of the opinion that the current technological progress in digital image processing and the explosion of imagery will have the same profound impact on the visual culture as did the visual innovations of the Renaissance (may I refer to my ideas about the topic of death of photography: a view now widely supported by many historians of art, but not by the internet community: the vox populi wins of course).
If this new Canon camera will materialize it should send shockwaves through the world of imaging, Hasselblad and Leica S2 owners might become nervous and Leica M designers would say with a smile: we were on the right track when we decided against a low pass filter. The PC manufacturers and external disc suppliers would see new sales coming: how to cope with such large image files and its backups?
The burning question would also be: are we able to get better pictures? The answer was given long ago by H C-B: pictures taken in 1850 look identical to pictures taken in 1990. Would it be different in 2012?

Another interesting aspect is the news, reported in the BJP, that Panasonic is now making lenses that do not have the Leica brand name, the reason being that Panasonic attempts to produce high quality images by lessening the demands on the optical design and improving the software processing the image. This division of labor I have predicted long ago when I noted that the mathematics required for optical lens correction could also be integrated in the post-processing software when the power of the hardware does allow this. Presumably Panasonic now thinks it has the stuff in house and does no longer need to follow the stringent demands from the Leica company. It also points to another strategy: lenses can be made with characteristics that are beyond the current state of the art of optical design but the inevitable degradation effects can be compensated by software. Why use aspheric surfaces to combat spherical aberration or why use expensive glass to correct chromatic aberrations when software can do this easily and above all economically? We might see a new era in lens design too!

Rational behavior?

According to the book “Freakonomics” persons may be acting more rational than is often assumed. Still irrational behavior is commonplace, especially in the domain of consumer electronics, which now sadly includes the world of photography too. There is a huge debate under philosophers-of-art whether photography can be included in the world of art and whether digital photography can be classified as photography. Very interesting to read and a topic more photographers should be studying. It is al least more rewarding than reading the tons of speculative ideas about the coming generation of cameras, the future of the mirror-box in the design of cameras and the emergence of 40 Mp sensors in what once was designated as 35mm cameras or miniature cameras. A 24x36mm sensor surface with 40 millions pixels will have a square pixel size of 4.7 micrometers and a total pixel amount of approximately 7676 x 5117. With this pixel size the Nyquist frequency would be (in linepairs/mm) more than 100 lp/mm. There would hardly be a lens available that can record such resolution, which is close to microfilm capabilities in the chemical workflow.
If you would propose to working photographers (as I sometimes did) to use a low speed microfilm to exploit the capabilities of Leica lenses the classical comment would be that no one needs such high definition, that Tri-X is best suited for 35mm photography and that big enlargements are the domain of medium speed cameras anyhow. To be able to show the level of detail that can be captured with microfilm you need really big enlargements or sectional enlargements and a level of technique that is beyond the capabilities or interests of most photographers. The same holds true for that 40Mp sensor the whole photographic world seems to be craving for. Two years ago everyone was satisfied and very happy with a 20 Mp sensor and I am sure that as soon as the 40 Mp sensor has materialized in some camera-body the cry will be for a 80 Mp sensor. Bigger is always better and more is even better than less.
This trend to ask for more in every area is irrational because it serves no purpose and actually makes things worse: you need to invest very heavily in precision manufacture, improved techniques (tripod etc) for a very limited improvement in picture quality (actually even a degradation in picture quality). A sensor with 40 million pixels based on a Bayer (three color pattern) makes no sense. The very fact that people are demanding this, shows the level of irrationality. If you would ask if someone would be interested in investing 100% more energy for a 5% more performance anyone would decline such an offer. But in the world of consumer digital imaging, rational thinking is scarce.

Mirrorless cameras

If Barnack would look at current trends he would undoubtedly conclude that his idea of a mirrorless rangefinder camera was sound and a valid proposition for a full century now. Current hype has it that the mirrorless camera (or EVIL) will kill the mirrorbox. The surge of the modern EVIl camera has a historical parallel: in the period from 1960 to 1990 the compact viewfinder camera (so called point and shoot cameras) threatened the existence of the SLR camera type. The compact camera had AF long before the SLR incorporated this feature and the cameras were indeed very compact and extremely potent with high quality lenses, not unlike the current MFT models. The truth is that the compact AF viewfinder camera disappeared and the SLR evolved into the dSLR. Presumably the MFT and EVIl models are the technological successors of the classic compact viewfinder camera. The fact that Nikon seems to jump the bandwagon of the mirrorless cameras is in my view an indication of lack of ingenuity and innovation in the Nikon company, the leaked specs of this camera type are not breathtaking to say the least.
Much more interesting is the silence of Canon on all fronts. I assume that Canon is taking a very in-depth investigation into the future of digital imaging and will adjust their line-up accordingly. The future of digital imaging is as insecure and unpredictable as the current economic crisis and Canon is possible looking at both: one is dependent on the other.

Fujifilm X100

There are now about fifty stories published about the Fujifilm X100 and many of these complain about the lack of speed of the AF and the confusion about the many menu options combined in several layers of menu choices. These comments are predictable: they are made about almost every camera and this type of remark is the result of being able to use a camera for a very short time and a focus on easily identifiable characteristics. One might assume that a mountain bike is a very simple and transparent technical product: but to get the most out of a bike one needs weeks of intensive use to know what the bike is capable of. I have the view that one needs a longer period of use to get really acquainted with a camera and detect its true use and character.
I am writing the test report now and it will be published in a week, but some remarks may be of interest here: when you select the camera options such that they reflect your picture taking style, the cumbersome menu choices are less relevant and to be honest with the speed of the AF I have noted that the AF is sluggish only when the objects are moving fast and to inject some relativism into the discussion it has to be said that manual focusing with the M9 is not easier when the objects are moving fast: ever tried to focus quickly and accurately on a fast moving cat that jumps in all directions?
The designers of the X100 have looked very carefully at the Leica camera, specifically at the M8, I assume. The M8 has a sensor size slightly bigger than the APS-C in the X100, but with a comparable amount of pixels (ten million). The X100 lens has the classical specifications of the most used optics for street photography, a 35mm f/2 lens. The most innovative feature is the hybrid viewfinder which is indeed amazingly useful: you might see the satisfaction on the Fuji engineers muttering that this approach might signal the future of the rangefinder technique. Leica would of course counter that they have different and better ideas, but for this moment the Fujifilm statement stands.
The performance of the lens is very intriguing: it closely resembles the quality of the very classical and famous Summicron 35mm, version 4, the last pre-aspherical design. You might wonder why Fujifilm needs such an advanced design to emulate a classical lens, but presumably this is on purpose. The X100 has an integrated leaf-shutter with a reduced top-speed when the lens is wide open and ND filters are necessary to reduce the amount of light falling on the sensor. This construction does not allow the possible inclusion of interchangeable lenses. In this sense the X100 might be seen as a modern interpretation of the very classical Konica Hexar with fixed 35mm/f2 lens, in those days a competitor of the Leica M camera for street photography.
Picture taking is a process that is mainly located between the ears and behind the eyes of the operator. The added value of the camera technology is of course noticeable and influences the resulting style of picture taking. The X100 has a large amount of unnecessary features, but basically it is a very pleasurable camera to use and at least one of the few models on the market which invite you to think about the rangefinder future.