Olympus E-1: the Barnack camera for the digital era?
Intro
Let us review the stage of photography at the beginning of the 20th century, the period when O. Barnack was thinking about the design for a totally new photographic instrument. The Leica camera, as we know now, broke every rule that existed in the photographic community of that time. Cameras and negatives were big, the engineering was quite rough, the optical quality of the lenses was acceptable, given the contact printing if that period, where enlargements were not required. The Leica was small, of exquisite engineering, and required new lenses, designed specifically for the small format of the negatives that needed big enlargements to get decent sized prints. In the beginning no one believed that this 'toy' could seriously challenge the established conventions and techniques.
The totally new and daring concept of the Leica camera created its own Leica technique and technology followed quickly to support the Leica approach.
For almost a century the Leica technique dominated photography and the 35mm format became the yardstick for photographic quality and style. Of course, big formats lingered on as a niche product and filled the need for superior quality in tonal reproduction and detail richness.
Olympus OM system
The Olympus designers created the OM SLR system when the SLR systems were booming business around 1970. It was their stated goal to emulate the quality and compactness of the Leica M-system in an SLR guise. At first, they called their camera the M1, but Leitz protested as this designation was to close to the M-series. Thus the OM, Olympus M system was born. The company created a series of classical cameras, ending with the OM4, and an accompanying lens line of very good quality. Geoffrey Crawley, of BJP fame, to check whether the high praise of users had substance, specifically tested the Zuiko 1.8/50mm lens. GC did indeed conclude that the Olympus lenses were small and possessed excellent image quality. The OM series attracted serious attention from professionals, but never became a photographic workhorse like the Nikon and Canon cameras.
Olympus shared the almost fanatical dedication to engineering quality and optical excellence with the Leitz company and it may be no coincidence that both firms were heavily involved in the manufacture of microscopes and other scientific instruments.
Olympus E-1 system
As soon as it became evident that the solid state capturing devices could and would overrun the silver halide grains as the premium means for recording images, Olympus saw for them a chance to move forward in a largely uncharted land and create a pole position that they could not create in the film-based photographic world. Olympus became famous for the first advertising campaign to declare the death of film. As with the original Barnack camera, the Olympus designers and engineers started from scratch and created a camera system that was focussed on a small format chip and needed dedicated lenses to exploit the inherent image quality of the small sized sensor. The 5 MP chip was selected as being able to compete with and even outclass the 35mm format. Remember that Barnack and Berek in their days considered that one MP on the 24 x 36mm sized negative could compete with the much bigger negative sizes of conventional cameras.
The lenses for the E-system are indeed outstandingly good. The published MTF graphs show a contrast transfer of more than 70% for 60 lp/mm. This is much better than what (as example) current Leica lenses in the same class of focal lengths can deliver.
Olympus however made a marketing fault (in my view) to focus excessive attention to the telecentric nature of the lensdesigns. There is some telecentricity in their lenses, but that does not alone explain the high-class image quality. It is the design concept of the optical layout and the mechanical engineering of the mount that does the trick. Just plain solid optical design.
New myths are starting to be born.
Every age and period has a load of myths to deal with. And this digital age is no exception.
In the Barnack period the small size of the negative was seen as a problem and as was the need for new optical designs. Now we have the same situation. The pixel race may be over, according to some commentators, but a 5MP chip is no longer considered as a serious proposition. One needs at least 8MP or even more and a camera is now seen as a professional tool only when the pixel count starts at 10 MP. We are falling here in the same trap as the photographers in Barnack's time. Bigger is not better!
Follow this simple calculation: the Kodak CCD in the E-1 has a pixel size of 6.8 micron, a value that allows for the clean definition of 80 lp/mm. A 35mm negative has a much larger area, but when I scan this negative with the Nikon Coolscan 5000 ED, I get 4000 pixels per inch, which happens to be a pixel size of 6.35 micron. The final resolution is almost identical. A motive captured on 35mm film and on the Kodak chip will have the same definition of fine detail, when recorded with a lens of equal quality.
A larger sensor area will often have a larger size of the pixel and therefore the resolution will be lower (Nyquist calculation). If I put a lower quality lens in front of the larger sensor, I will degrade the final performance with another factor. A larger sensor may allow for a bigger print size, but that amounts to the same argument as heard in filmbased photography: a negative from a Hasselblad can be enlarged more with the same quality than can a negative from a Leica. This is true, even when we all agree that the optical performance of the Leica lens is better.
The superior image quality of the Olympus dedicated lenses for the small Kodak sensor is based on physical size. Compare the size of a Leica lens with the negative area and do the same with the size of the Olympus lens with the sensor area. Then the Olympus lens is relatively speaking a huge lens. The Zuiko 2/50mm lens (the 35mm equivalent would be a 2/100mm) has a length of 71mm and a diameter of 61mm. That represents a volume of 207390 mm^3. The Apo-Summicron-M 2/90mm has a length of 78mm and a diameter of 64mm. That equals a volume of 250798 mm^3. Roughly both lenses have the same volume, but the diameter the sensor of the Kodak chip is 22mm and the diameter of the 35mm negative is 43mm, twice as large. Many lens designers also complain about the restrictions imposed by the diameter of the bayonet flange because the light rays have to be squeezed through a relative small hole in the light path.
Camera lenses for 35mm format have always been constrained by the possible maximum size of the mount and optical designers all over the world will remark that the size limitations will also inhibit the exploitation of maximum optical performance. If one could design a lens for optimum performance without size restrictions, the quality could be much higher. One notable lens system that tried to create lenses without size limitation was the Zeiss line for the Contarex, a system that stayed for a long time the yardstick for optical excellence.
A camera system with small sensors will offer the optical designer much needed leeway for designing lenses with a larger physical size to boost performance to a level that ius not possible in 35mm lens systems. The APS format was originally proposed to enable smaller cameras with relatively bigger lenses to optimize the potential of the film emulsions. The APS format is now the most widely employed sensor size in modern D-SLR's. This size is selected to give users of classical lens systems a compatibility bonus, as the lens factor of 1.3 to 1.6 is just acceptable to most users. The size is not necessarily selected for better imagery.
It may be the case that a modern D-SLR with 10 to 12 MP and a lens designed for 35mm photography or a lens adapted to the APS capture circle will not produce better results than the combination of a small sensor and a dedicated lens design, as for example the E-system of Olympus.
My own comparison tests of the Olympus performance and the performance of the Epson R-D1 with a superb Leica lens have as conclusion that he above statement is not yet falsified. I am doing more tests to get a broader view of the matter. Read the test!
Barnack was right against all odds 100 years ago. Could the design team of Mr Asakura repeat the genius of Barnack? (2005)