Haruo Ogawa, President of the Imaging Business Group and Executive Managing Officer at Olympus Corporation.

When we attended CP+ last month in Yokohama, Japan we sat down with senior executives from several major camera and lens manufacturers. Among them was Haruo Ogawa, President of the Imaging Business Group and Executive Managing Officer at Olympus Corporation. Among other things we spoke to Mr Ogawa about the current and future direction of Micro Four Thirds and the challenges of introducing 4K video. 


How has the OM-D E-M5 II been received?

It’s early days but opinions from press and professional photographers have been favorable.

You’ve added a new 40MP high-res shot mode to the OM-D E-M5 II, are your customers asking for higher resolution in conventional shooting modes?

For normal shooting no, we have not received any requests for resolution higher than 16MP. On the contrary, our customers are more than satisfied by the picture quality of the OM-D system. But we did receive requests from studio photographers who needed higher resolution. Because of the size of the sensor, we didn’t go in the direction of higher pixel count, so to meet the demands of those photographers we developed 40MP high-res shot mode using pixel shift. 

We have received some initial feedback from these studio professionals and they’re telling us that image quality - especially at the corners - in 40MP high-res mode is better than full-frame DSLRs. 

16MP is a bit low compared to some of the competition from 1-inch, APS-C and full-frame sensors. Are you interested in increasing resolution?

We are considering higher pixel counts in future models, but more importantly we need to make sure that image quality is maintained, in terms of noise levels and other parameters. We will gradually increase pixel count in the future, while maintaining picture quality. Pixel count is just one factor in image quality. Lens quality, the image processor performance and sensor performance are all very important. 

The Olympus OM-D E-M5 II is the successor to the original OM-D E-M5, and brings a host of additions, including an innovative 40MP multi-shot mode and faster continuous shooting and autofocus. 

What kind of photographers do you see taking advantage of high-res shot mode?

You have to use a tripod with this mode, so the main users will be studio photographers. But nature and scenery, for still scenes, this mode is good [for that] too.

We’ve seen software solutions which stitch together multiple slightly shifted images to make a higher-resolution image. This kind of software automatically detects ghosting from moving scene elements and attempts to reduce it. Does your high-res shot mode attempt to do this too?

This is interesting technology - same as what is used for HDR - but we’re not currently studying it for the high-res shot mode. 

In the future, do you think high-res shot mode could be speeded up enough that the function could be used in hand-held shooting?

Yes this is technically possible in the future. It depends on the performance of the camera’s processing engine, so as the engine performance improves, then yes - it should be possible.

In theory, if a Panasonic lens with optical image stabilization was attached, super high-res mode might be useable for hand-held shooting in some situations, but lens-based image stabilization is disabled in this mode. Is there a technical reason why?

I don’t know - I haven’t tried it!

High-res shot mode works by shifting the OM-D E-M5 II's sensor multiple times in minute increments to create a final 40MP image. Static scene elements are rendered with excellent detail but any movement at all (like the water in the foreground in this image) results in unpleasant artifacts.  

As a camera manufacturer, are you more interested now in enthusiast photographers or beginners?

We’d like to cater to a broad spectrum of users, at both the low and high-end. For beginners we have the PEN series and the OM-D E-M10 and for high-end users we have the E-M5 II and E-M1. 

Do you see a difference in the lens attachment rate between cameras like the E-M10 and the E-M1?

Obviously the lens attachment rate is higher for the high-end cameras, but entry-level users are beginning to realize the benefits of changing lenses thanks to our affordable accessories like our lens converter attachments and our body-cap lens. Our 45mm F1.8 is very popular with our entry-level users because of its high quality and affordable price.

Can you give us some idea of your lens roadmap?

Our priority first of all is the pro and premium series of lenses for enthusiast and professional photographers. We also want to add some more reasonably-priced lenses for entry-level users, similar to the 45mm F1.8 - we’re looking into this now. 

When you create a camera like the E-M1, what are the competitive cameras that you benchmark against?

We don’t benchmark against any particular specific model. We know the benefits of full-frame DSLRs but one area where we think that Micro Four Thirds has an advantage is lens performance, because full-frame lenses have to cover a larger area. We can create higher image quality. So we put the emphasis on portability and image quality. That’s why we don’t benchmark against any specific models.

We are very keen on lens quality. If we were to develop the same quality lenses for full-frame cameras, the size would be four times bigger. But if you compare our lenses against [lenses designed for full-frame mirrorless cameras] you’ll see that there isn’t a 4X difference in size. That means that there’s some compromise in lens performance [in those lenses]. 

Olympus has been working on Micro Four Thirds for many years now. If you could go back to the beginning of the process is there anything you’d do differently?

No, we’d probably still choose to create the Micro Four Thirds system. We have 80 years of camera making history. We started our business in 1936 and our consistent policy since then has been to make our cameras compact and lightweight. For example [with M43] a 600mm equivalent lens can be attached to a camera and used hand-held. This would not be possible with a full-frame model.

Olympus's forthcoming 300mm F4 lens offers a focal length equivalent to 600mm. Although not as capable of the same things as a 600mm F4 would be for full-frame, it's highly portable by comparison. 

Imagine shooting wildlife on safari with a full-frame 600mm, where you might not have time to set up a tripod, etc. We also hear from photojournalists in conflict zones who need a very compact, light system. Even for portrait shooting, models are more nervous with big cameras. 

We don’t dismiss full-frame models, of course but there are [types of photography for which] they’re not appropriate. So if I could go back ten years I wouldn’t change anything.

Do you know what portion of your customers are professional photographers?

I don’t have exact numbers but after launching the OM-D the number has increased, especially in the documentary field. Documentary photographers tend to use their cameras in tough environments.

Are you looking to develop on-sensor phase-detection autofocus in future models?

Before Micro Four Thirds we had Four Thirds. Lenses designed for Four Thirds require phase-detection autofocus [in the camera]. So in order for these lenses to still be utilized [in M43 cameras], we will continue to develop phase-detection AF. For this reason, we’d like to improve AF speed and precision. E-M1 firmware version 3.0 increases continuous shooting with autofocus tracking to 9 frames per second. This is one result of our improvements to phase-detection AF.

Do you think on-sensor phase-detection will ever fully replace contrast-detection in mirrorless interchangeable lens cameras?

No. Because there are some areas in which contrast-detection has advantages. One example is autofocus towards the edges of the image area. Another is our face and eye-detection focus feature. 

Do you think the best solution is a combination of the two types of autofocus?

Yes. Phase-detection AF cannot incorporate features like face and eye-detection, so I believe that the best solution is contrast-detection plus phase-detection. [Editor's note: while traditional phase-detection AF modules cannot by themselves detect faces and eyes, secondary image sensors, or the main imaging sensor itself in mirrorless cameras, can 'tell' the camera which phase-detection points to use for focus. Hence, technically, contrast-detection AF is not needed for face or eye-detection] 

Let’s talk about the Air. Right now it’s only going to be available in Japan - are you thinking of expanding into other markets?

We’re in the test stage now, in terms of sales. So we’ll see how Japanese customers react, and that will help us decide whether or not to expand sales into other markets.

The new 'Air' consists of a 16MP M43 sensor and lens mount inside a module that can be attached to various things, and 'hacked' by developers. Olympus has introduced the Air to Japan but it remains to be seen whether it will become available anywhere else in the world.  

In terms of development tools for the Air, do you plan to make all of the camera’s functions accessible to developers - for example autofocus control, image processing, etc.?

We allow access to some functions, but we have not yet decided whether we will allow access to all functions. The first thing we want to know is how this concept will be accepted by developers. Depending on how they react, we’ll decide on how we want to improve. 

What is the biggest challenge that faces Olympus as a camera manufacturer?

There are some emerging trends in the camera market - some manufacturers are making full-frame DSLRs for example and some are introducing 4K movie recording, but our approach is to provide cameras which are compact and lightweight. Explaining why our direction is important, and to whom is a big challenge. 

Are you thinking of adding 4K video capture in future models?

4K movie places a large burden on the image processor, so there are tradeoffs. This time we prioritized image stabilization, but in the future, when processor performance increases we can provide both.

Mr. Ogawa - how long have you worked at Olympus?

34 years. I started in R&D, and my first job was working on Zuiko lenses. After moving into imaging my first job was working on the Stylus series. On DSLRs I worked on the E-410, E-510 and E-3. 

What are you most proud of?

I am proud of all of it, but mostly the development of the first PEN at a time when there were no mirrorless cameras in the market. It was a really original idea. The style, and the design of the E-P1. Until then, the DSLR market consisted of around 90% male customers, mostly over the age of 50, but with the introduction of the E-P1 we attracted young, and female customers to the interchangeable lens camera market. 


Editor's note:

Two phrases cropped up a lot during our interview with Mr. Ogawa - 'full frame' and 'image processor'. It's interesting that Olympus seems more preoccupied with its full-frame competition than APS-C but not altogether surprising - that's where the biggest gap exists in terms of critical image quality. The OM-D E-M5 and E-M1 punch above their weight when compared against their APS-C peers but neither can really challenge full-frame mirrorless and DSLR cameras in terms of resolution, depth-of-field control, and low light performance. High-res shot mode is a nice feature, and a magic 'bonus' 40MP output mode will appeal to a lot of people. The problem is that because it is limited to entirely static subjects and tripod-mounted shooting, it will only be genuinely useful (for now) to a very narrow constituency of photographers.

Where Olympus can definitely compete is in terms of lenses, and sheer weight of camera functions. I'm not surprised that the OM-D series has proven popular among documentary photographers, and Mr. Ogawa is right that there are some situations where larger full-frame system cameras and lenses are not appropriate. The quality of Olympus's recent 'Pro' lenses and primes is excellent, too. You can nitpick about aperture equivalence all you want, but the quality and versatility of a lens like the 40-150mm F2.8 Pro is hard to argue with. The OM-D E-M5 and E-M1 are incredibly feature-rich, too - almost to a fault. Deep customization and a huge number of physical control points means that both cameras can be set up to cater to even the most esoteric of ergonomic preferences. 

We were encouraged to be told that Olympus is 'considering higher pixel counts' in future OM-D models. I don't completely believe Mr Ogawa when he told us that Olympus has received 'no requests' for resolution higher than 16MP but it makes sense that the company is being cautious. To reap the full benefit of more pixels on a comparatively small sensor, though, lenses will have to be up to the challenge, and we have no doubt they will after seeing the performance of some of the latest and greatest offerings.

On this topic, Mr. Ogawa's comments that micro four thirds lenses have an advantage in terms of performance was particularly interesting. On the one hand, creating a smaller image circle may allow them to focus much of the (size, weight, and complexity of the) optics on performance, while keeping the lens relatively small. However, the micro four thirds format is challenged by the requirement of higher resolving power lenses simply to maintain the same effective total image resolution as a larger sensor counterpart, for any given field-of-view that is. Furthermore, micro four thirds lenses become diffraction limited earlier when comparing equivalent resolution cameras, as diffraction is correlated with equivalent, not absolute, aperture. What this means is that rather than any one system having a clear benefit or disadvantage, lenses for different formats face different sets of challenges.

Which brings us to the second much-used phrase in this interview - 'image processor'. The fact is that headline OM-D features like 10fps maximum shooting, complex multi-axis image stabilization and multi-shot high res etc., require a lot of processing power. Overtaxed processors get hot (a real problem in very small cameras) and adding more powerful ones costs money. It's clear that the processing challenge in general is one reason why Olympus has yet to add a 4K video mode to its OM-D lineup, too.

We asked Mr. Ogawa about the biggest challenges facing Olympus as a camera manufacturer and his answer was revealing. He didn't point to any specific technical hurdles, or even to broader market conditions, but to communication - the challenge of educating consumers about the benefits of M43 compared, specifically, to full-frame.