micro four-thirds

translation of a dcwatch interview with

ogawa harou of olympus, japan

translated by 'hokuto'


Just last week Olympus and Panasonic announced a new 'standard' for a new range of smaller DSLR cameras they are to produce - micro Four-Thirds. DCWatch, a Japanese based website were sharp off the mark to interview Ogawa Harou of Olympus Imaging, Japan about the new standard and the cameras likely to follow. The interview appeared on the DCWatch website and has been translated by 'Hokuto' of the Four-ThirdsPhoto Forum. Hokuto very kindly agreed to let me publish his work on this site. (Hokuto accepts no responsibilty for its accuracy or content).


Last week, Olympus and Panasonic announced 'Micro Four Thirds' as an extension of the Four Thirds Standard. By reducing the flange-back distance to 1/2 that of previous standard, and reducing the mount diameter by 6mm, it represents a branch out from the Four Thirds standard (which is being continued as before), that makes it possible to reduce the size of camera and exchangeable lenses. We spoke to the head of Olympus Imaging?s SLR division, Ogawa Haruo (interviewer is Honda Masakazu, of DCWatch).

The Four Thirds Ideal does not change with Micro Four Thirds.

Before entering the main theme of our interview, Mr. Ogawa said he wanted to appeal once more to the strengths of the Four Thirds standard. "We've heard rumors that this announcement means Olympus is going to abandon 4/3. But we are already scheduling to announce multiple 4/3 bodies within this year, and we have absolutely no intention of abandoning the 4/3 standard. So here, I first of all want people to stop thinking that the announcement of Micro Four Thirds [hereinafter I'll abbreviate this as M4/3 -- trans.] means that 4/3 fans have been betrayed."

Mr. Ogawa was originally a technician and spent 20 years occupied in research. During his time in the labs, around 1996 research was progressing quite nearby regarding the issue of 'the optimum SLR system for digital cameras.'

'During the film era with manual-focus cameras, we at one time abandoned the OM mount and withdrew from the world of SLRs. So we thought, if we didn't consider the past, and started out from zero, what kind of design would produce the ultimate balance of high image quality (equivalent to the 35mm SLR system) and portability? It was in response to that question that the 4/3 system was born.'

Ogawa continued with conviction; 'At that point back in 1996, we were already thinking that lenses needed a minimum MTF of around 200 lines, that's about five times the resolution necessary for film lenses, which were okay with 50. Naturally, there were also issues about light falloff at the edges, degradation of resolution, moire, and false color [CA, etc.] We said, 'wow, that's a tall order.' Conventional thinking and common sense won't be enough here. 4/3 was announced in September, 2003, but that kind of step-by-step research lay in the background to that announcement. In short, 4/3 was our answer to the question of what kind of format would be optimum for producing sufficient resolution while preserving portability, so we're not about to throw that out and redefine another format.'

He added, 'I want to say something about the issue of telecentricity. It's often said that the reason for seeking telecentricity is because light won't reach the bottom of 'deep wells' if it doesn't enter at the perpendicular. But in fact, there are a lot of other issues involved. Will (the sensor) resolve all the way to the borders? Can you aggressively create the images you want? For example, the depiction of the deep sky color sometimes called 'Olympus blue.' We can achieve that kind of color because we are bring the ideal light to the sensor.'

'We often hear the knee-jerk response that 'image quality is bad because the sensor is small,' but excellent images cannot be produced except by excellent imaging produced by excellent lenses. Lots of manufacturers are heading toward 35mm full-frame sensors, but they have to be facing considerable issues, such as light falloff at the peripheries, MTF degradation, and chromatic aberration.'

'Needless to say, if you produced lenses for full-sized sensors that were 4x the size of 4/3 lenses, you could produce the same image results with full-sized sensors. And there are opinions to the effect that some of the shortcomings of lenses can be made up for by electronic means. But at Olympus we didn't want to digitalize the SLR by such short-sighted tactics.'

"Liveview" a pivotal role in the standard.

DCW QUESTION: A lot of people, myself included, have said to Olympus that we'd like 'another 4/3', one that made greater use of the compact size of the sensor and allowed a shorter back-focus distance [note: "back focus" here means the distance from the rear-most lens surface to the imaging (sensor) surface; according to Wikipedia Japan, it's often confused with "flange-back" distance, but they're not the same] and utilized EVF or liveview, or a rangefinder. It's likely that those kinds of opinions were heard within Olympus as well, so how long have you been working to cook up the M4/3 standard?'

OGAWA: 'First of all, it was necessary for us to produce high-performance bodies that would satisfy 4/3 fans. In addition, we introduced bodies with IS and smaller bodies that made greater use of the compact sensor size, in that way producng a general lineup of models for the 4/3 system. Only then could we set our hand to the M4/3 system.'

'But producing a concept and actually bringing it to market are two completely different things. Even with the issue of shortening the back-focus, we made sure that the new standard would support the stock of previous 4/3 lenses. Then again, by shortening the back-focus distance, it was necessary to eliminate the mirror box, so it was important to consider the timing when substitute products with fully practical performance [I assume he means EVF, etc.] would be available.'

DCW QUESTION: This time the announcement was a joint announcement with Panasonic, but what new was produced in your joint talks?

OGAWA: 'M4/3 was something that was brought to fruition jointly with Panasonic. Panasonic has been in the 4/3 camp since 2003, but even back then this kind of talk was being banded about. Both of us were asking, "Will a real camera come out of this?" and "Can we make optimum use of 4/3?", and from those talks the M4/3 was born.'

'To predicate a system on the removal of the mirror box is much easier to say than to accomplish in fact. First of all, things like liveview and EVF have to be able to function practically or the system is useless. And since it was necessary to make autofocus of the contrast detection type, it was necessary to proceed jointly on the development of the LiveMOS sensor. But 4/3 had implied that kind of development from the beginning. That?s why we already had it in our minds to support lenses for contrast AF by updating the firmware.'

DCW QUESTION: In your press release materials you use two different expressions, "digital single-lens system" and "digital single-lens reflex system," but is this because with the flange-back of only 20mm it makes the 'reflex' an impossiblity? Based on size calculations alone, it would seem that if you eliminated the dust removal device, it would be barely possible to introduce a swinging mechanism?

OGAWA: 'It's probably impossible to put in a mirror under the present [M4/3] standards. I wouldn't say "absolutely" but the dust reduction system (even though it's not actually included as a part of the standard) is implictly understood to be there. Further, although the flange-back is said to be about 1/2 the previous value, thus around 20mm, the precise measurements differ. That's because the mechanical dimensions are not made public to parties outside the 4/3 consortium.'

Shooting styles with high degree of freedom.

DCW QUESTION: I wonder how much more compact a system M4/3 can be compared to the current 4/3 standard. For example, in the conventional, standard zoom range?

OGAWA: 'The Four Thirds standard utilized a long back-focus distance in order to enhance the telecentricity in all focal lengths, from ultra-wide to ultra-telephoto. Also, an essential benefit of the FourThirds system is that it makes it possible to reduce the size of telephoto lenses. But wide-angle lenses unfortunately have to be made larger. However, there was the potential that advances in lens technology would make it possible to solve that issue.'

'In fact, the ZD ED 9-18mm f4-5.6 lens announced just recently would have been unthinkable before. It was only because it became possible to mass-produce an aspherical lens with enormous variable ratio that it was possible to make it a compact lens without sacrificing image quality.'

The question of "how small" just depends on the design and what kind of lens to make it. Compactness is important, of course, but after all, one chooses an interchangeable lens camera because one wants things like defocuisng characteristics (bokeh), the overall way the image comes together, depth of color, and so on. If the lens isn't good, the image quality won't be good. More than simply making the lens compact, it's necessary to ensure that fundamental imaging quality is high on the optical level right out to the edges, and without compensation using digital technology.'

DCW QUESTION: Yes, that's all very clear, but doesn't the consumer also want to know, even roughly, what kind of lens system - and body system "it's" going to be?

OGAWA: 'Both body and lens will be smaller than the (current) FourThirds system. That goes without saying. Also, since the mirror box will be eliminated, the viewfinder can be freely positioned anywhere, so it should be possible to propose shooting styles offering a degree of freedom unattainable before. Of course, this means the potential to develop displays that show shooting information on the EVF and liveview.'

DCW QUESTION: Certainly, it should be possible to overlay on EVF or liveview displays a wealth of information that would be impossible on conventional optical viewfinders. That's an important point to look forward to, but on the other hand, one can't escape the negative impressions regarding the speed of response [for immediate shooting]. Can we surmise that you have announced the M4/3 standard at this point because you are able to now see shooting response with an EVF approaching that of an SLR system?

OGAWA: 'Practical use of liveview has been considerably improved with the E-420, and if we continue improving that system, we anticipate it will be possible to use it at a level without significant problems. Technical development continues to advance. However, if one asks whether it is currently on a level with current SLRs, it?s not there yet, no. We want to develop M4/3 over a long span of time, so please take things on a bit more of a long-term view.'

No Problems with Telecentricity?

DCW QUESTION: The mount diameter has become smaller, but will that cause any problems with telecentricity?

OGAWA: 'The original 4/3 standard used a very large mount diameter, so reducing it by 6mm will not cause any problems. The concept of introducing light rays at a perpendicular angle to the sensor will carry on into the M4/3 system.'

DCW QUESTION: "Back-illuminated" imaging sensors are steadily becoming more realistic, and with the introduction of imaging processing technology that eliminate low-pass filters, it seems that telecentricity won?t be so necessary in the future as it is now. If so, then the length of wide angle lenses will become shorter, so it should be possible to make an ultra-compact system that makes the most of the benefits of the short-back-focus distance. I really thought that the M4/3 was going to be seated firmly in order to take advantage of that kind of technological direction, but . . .

OGAWA: 'Certainly, it may turn out in future that telecentricity is not required so much as it is today. But at present, technological development hasn't progressed to that point, so we continue to believe that telecentricity is important. Telecentricity in the 4/3 system does not exist merely as numbers sought in lens design, but represent a common consciousness among the manufacturers participating in the 4/3 consortium. We don't know what the future may hold, but it doesn't make sense to make lenses for use only in cameras that use a certain kind of sensor, so for the present we continue to design lenses with an important emphasis placed on telecentricity.'

DCW QUESTION: The M4/3 supposes a compact imaging sensor with interchangeable lenses that have short back-focus. Assuming an EVF, the borderline with video cameras will become more and more ambiguous. Do you also have a consciousness of including motion video capture? For example, it you had a camera that not just allowed conventional "video" shooting, but was specially oriented toward shooting with interchangeable lenses on a par with "movie making" - that would represent quite an interesting development.

OGAWA: 'We have our eye on moving pictures, yes. But there would be nothing interesting if it were merely the addition of a video function [like ordinary P/S cameras]. As you say, [true] movies are the continuous shooting of still images, so we think it would be interesting to have a camera that shot movies with the same sensibility as our own still-picture images. In that, we can take advantage of the excellent 4/3 lenses. Afterwards, it would be nice is we had functions that linked to SNS sites and movie upload sites.'

DCW QUESTION: Have you thought of building a camera that uses the 4/3 sensor, but without interchangeable lenses?

OGAWA: 'We haven't considered it in particular. Of course, there may be talk of using a 4/3 standard sensor in a compact camera, but we aren't thinking of defining that kind of product as part of the M4/3 standard.'

DCW QUESTION: Two additional pins have been added between the lens and body, but what are they going to be used for?

OGAWA: 'They'll be used to enhance the control characteristics of the lens. They'll allow the lens to extend more rapidly and more accurately than before.'

DCW QUESTION: Can we understand it to mean that the focus positionng accuracy will be increased by two bits?

OGAWA: 'Well, one can pretty well estimate how behavior will change by increasing control bits by 2. It will mean greater accuracy.'

DCW QUESTION: The 4/3 standard started as an open lens-mount standard, but how will the M4/3 standard be treated?

OGAWA: 'Already five years have passed since we first announced FourThirds, and our corporate stance toward digital cameras has become settled. Panasonic aggressively stepped forward and so we jointly made the M4/3 standard. But when many items have to be negotiated between individual corporations, the pace of advance slows down, so we can't do this with any company that isn't aggressive about it. In that sense, it's different from before. We're still canvassing for partners, but it won't be a completely open standard.'

DCW QUESTION: With the M4/3 system it should be possible to use Leica M-mount lenses through a converter. I'm excited about the potential of using those old interchangeable rangefinder types of lenses, but will it be possible?

OGAWA: 'As an interchangeable-lens camera system, [M4/3] will have the shorgest flange-back distance, so it should be possible. I personally hope that it becomes possible. As a lens-mount standard, some problems may occur in guaranteeing operation, so it may not be possible to support it from the corporate perspective, but I do hope that it comes about.'

END INTERVIEW/TRANSLATION.


HOME | INTRO | BOOKS | CIRCLE | QUEST | MEMBERS | GALLERY

CLASSIC | DIGITAL | MICRO FT | CORNUCOPIA | NEWS | SITE A-Z | SITE TOOLS


Posted 16/08/2008 Copyright © 2008 John Foster & Hokutu