We recently returned from the CP+ show in Yokohama, Japan, where we sat down with executives from most of the major camera and lens manufacturers to get their insights and opinions on the challenges facing their companies and the market as a whole. One of them was Masaya Maeda - Managing Director and Chief Executive, Image Communication Products Operations at Canon. 

Masaya Maeda - Managing Director and Chief Executive, Image Communication Products Operations at Canon, photographed at CP+ in February 2015.

Canon offered a maximum pixel count of ~22MP for seven years before the launch of the EOS 5DS and 5DS R. Why did it take that long to increase resolution?

I wasn't actually aware that it had been seven years! The biggest reason it took this long to get to 50MP is the targets we set for pixel-level quality. Every year we sit down with photojournalists, and listen to their demands - what they need to do their job - and the biggest demand was that even if pixel count increased, actual pixel-level image quality would not decrease. 

Obviously at the same time as we were seeing demand for increased pixel-level quality we were also being asked for increased pixel-count. The reason the time finally came for us to release the EOS 5DS and 5DS R this year was that we found a means of satisfying the demand for increased pixel count without sacrificing quality. It took us that long to reach this level.

You must watch the industry very closely - what did you learn from watching Sony and Nikon go before you, in terms of offering nigher resolutions?

There's nothing in particular that we learned from Nikon or Sony, but as I said before, we had many demands from photographers all over the world not to sacrifice image quality, so that's what we placed emphasis on and our main priority [during the design process of the EOS 5DS and 5DS R] was to satisfy those photographers.  

The new Canon EOS 5DS and 5DS R offer 50MP resolution, with (in the case of the 5DS) and without (the EOS 5DS R) an anti-aliasing filter. 

Aside from pixel-level image quality, what were the main challenges Canon faced when creating these new cameras?

Of course with the increased pixel count comes more data, so there’s a requirement for increased processing to handle it. So we introduced the Digic 6 platform which is able to handle that amount of data. 

Because the pixel pitch is so similar we’d expect the 5DS and 5DS R to give similar pixel-level image quality as the EOS 7D Mark II, and therefore even better image quality at common viewing sizes. So why is the maximum ISO limit so low?

The users that we assume will be using the 5DS and 5DS R are professional-level users, whereas the EOS 7D Mark II is intended for a wider range of users. Users of the 5DS and 5DS R have much higher, stricter standards. So we set the ISO cap in accordance with what we felt that they would demand and require. 

Why is the EOS M3 not available in the American market?

I’m speaking for Canon Inc. The global headquarters. We offer a full lineup of products from compacts to mirrorless and DSLRs. We offer this full lineup to our sales and marketing divisions in each region. They decide which products they want to introduce into the market, so the decision not to launch the EOS M3 in America falls to Canon USA, not Canon Inc. I’m saying “sell it!” but it’s their decision whether they want to or not. 

In your opinion, what kind of mirrorless camera would sell best in the USA?

To be honest I don’t really know - I’m not that close to the US market so I can’t speak from first-hand experience. However I get the feeling that users in the US don’t really take a liking to small cameras. That’s just my sense. 

The 24MP EOS M3 is Canon's most enthusiast-oriented mirrorless camera yet, but it isn't coming to the US market. 

When you look at mirrorless cameras from your competitors, what scares you the most?

Looking at all of the mirrorless cameras out there, there’s nothing that really frightens us. 

The reason we ask is that mirrorless manufacturers are being forced to innovate a lot in areas like on-sensor phase detection autofocus and subject tracking etc. We wouldn’t be surprised if they catch up to DSLRs at some point.

We acknowledge and respect the fact that our competitors are innovating in the technology that they’re introducing to the market. But at the same time we are also innovating. Our efforts towards the development of mirrorless cameras are very serious. But to be honest when we’re looking at mirrorless cameras, and entry-level DSLR cameras, and high-end compact cameras, we don’t know which of those will become mainstream. 

So rather than looking at our competitor’s mirrorless cameras and regarding them as a threat, within Canon, our team who’s working on mirrorless should view the DSLR team as a threat. They should view the high-end compact cameras team as a threat. The threats that our mirrorless cameras team face aren’t from other companies, they’re from other divisions within our company.

A lot of people interpret the small lens lineup for EOS M as meaning that Canon isn’t serious about the system. Is that true?

Within the mirrorless segment we have every intention of launching more attractive lenses. 

Video is a major focus for Canon - would you rather that enthusiastic videographers invest in Cinema EOS, or your DSLRs?

That’s a difficult question - both! Again this is just my personal opinion but I believe that at the top-end we’re going to see a fusion with still images and video. At some point 8K will come for that reason, in the future there will need to be one product that will satisfy both needs.

When that hypothetical camera comes along - that camera which can satisfy the requirement for both stills and video - presumably it won’t have a mirror?

Maybe. 

How are you addressing the demand for compact, 4K-capable cameras like Sony's a7S and Panasonic's GH4?

Naturally I can’t give you details of our future plans but we are fully aware that such requirements exist in the market. We’re currently in the process of investigating, mainly to satisfy the needs of news media, and we have every intention of addressing this need in future products. 

One area where Canon has a competitive advantage is dual-pixel AF. Do you see Canon introducing this feature across your DSLR and mirrorless lineup?

It goes without saying that Dual-Pixel AF technology is very very effective, not only for stills but for movie shooting. For that reason we have every intention of taking advantage of this in our offerings to our users.

Canon's schematic of its Dual Pixel CMOS AF sensor structure. The top layer illustrates the light-gathering micro-lenses and conventional Bayer-type color filter array. The lower layer shows how each pixel is split into two photo-diodes, left and right, which are colored blue and red respectively. (Note that this does not indicate different color sensitivity.)

Dual-Pixel AF does not yet work for continuous predictive autofocus when shooting stills. Is this something you’re looking into?

Yes this is something we’re working on but it’s very difficult. The challenge is data. Processing that amount of data. If we were using a supercomputer it would be OK!

Is there room for an EOS-1D level product in Canon’s lineup, with a higher resolution?

This is something that is currently under consideration.

Hypothetically, if Sony made an APS-C or full-frame sensor that you considered to be the best possible sensor for a camera that you wanted to bring to market, would you use it?

If another company made a sensor that we believed to be truly the best quality, we would not hesitate to use it.

Your new 11-24mm is impressive, but expensive. If you could, would you sell it for less?

I’m using my own money to buy these lenses so I’d have to agree with that thought. I love the new 11-24mm. But it’s expensive! But one of the main components in that lens is a large, aspherical element which takes a long time to shape and polish.

We noticed that Canon has been granted a patent to use contrast-detection focus information in live view to automatically calibrate a DSLR for accurate phase-detection AF with an attached lens. Given that phase-detection autofocus is inherently less accurate than contrast-detection, is this something you are working on?

This is not something that I am aware of. We are granted a lot of patents. Even within the limited scope of patents related to combining or making use of contrast and phase-detection AF I know there are so many patents which we’ve filed. We file thousands and thousands of patents.

As a company last year we were number three, I think, in terms of US patents awarded. You should write a feature on patents! It’s a really interesting peek into the future. I really enjoyed writing them, and I enjoy reading about them.

How many patents do you have against your name?

Maybe… between one and two hundred. 

Quite often, one of the criticisms leveled at Canon is that the company is a little slow to produce innovation compared to competitors. There’s definitely a perception that Canon goes carefully, and slowly. Is this true? 

Personally, I think we’re slow as well. Every day I’m saying ‘speed up, make it faster!’. One of our themes now as a company is upon developing a new technology, to shorten the time between development and when that technology is introduced into a product. We need to shorten that time. That’s our goal. 


Editor's note:

As usual, the on-record portion of our interview with Mr. Maeda garnered some surprisingly open admissions, and some careful obfuscations. I didn't expect him to be quite so bold when I asked whether Canon would ever use a superior (hypothetically speaking, of course...) Sony APS-C or full-frame sensor, and I didn't entirely believe him when he said that in this (hypothetical, you understand...) eventuality Canon would buy it 'without hesitation'. 

What I am very sure of, after speaking with Mr. Maeda at CP+ is that he is driving the company to move faster, and more confidently. Canon's DSLR lineup might have been looking a little staid for some time, but recent releases like the EOS 7D Mark II and the new Rebel T6i and T6s are impressively innovative, and Canon's recent high-end lenses have been truly excellent. The 11-24mm, especially, was generating a huge amount of interest from both customers and competitors on the show floor at CP+.

The new EOS 5DS and 5DS R fulfill the promise that Mr. Maeda made to us at last year's CP+ that higher resolution sensors were on their way. Ironically though, given the amount of commonality with the older 5D III, they might count as Canon's most conservative camera releases for a while. Except for the EOS M3, that is. Despite Mr. Maeda's insistence that Canon is taking mirrorless very seriously, the M3 is not the camera we'd hoped it would be. Although undoubtedly the company's most enthusiast-focused mirrorless model yet, even in terms of bare specification it's no match for high-fliers like Sony's a6000 and a5100.

Mr. Maeda's firm insistence that Canon has every intention of releasing more 'attractive' lenses for mirrorless is encouraging though. Is a higher-end M-mount camera on the way? As Mr. Maeda might say: 'Maybe'. But will Canon USA say yes to it? Who knows. 


Note: This interview was conducted through an interpreter, and contained a mixture of on and off-record disclosures. As such, both questions and responses have been edited for clarity and readability where necessary.