Blog

On a recent trip to Thailand to visit Sony’s Chonburi factory, where all Sony ILC cameras and lenses are assembled, our tech editor Rishi Sanyal had the honor of sitting down with Kenji Tanaka, who is head of interchangeable lens cameras globally. Also joining the discussion was Daisuke Goh, product manager at Sony, and Matt Parnell, senior marketing communications manager at Sony.

Kenji Tanaka, Senior General Manager, Business Unit 1, Digital Imaging Group, Imaging Products and Solutions Sector. Photo: Rishi Sanyal

We discussed a number of topics, ranging from Sony’s intent to stimulate the market overall as opposed to segmenting it or supplanting any one manufacturer, to future sensor developments we can expect, to the role of computational photography and the internet in future cameras. It was an impressively open and honest discussion, and we thank everyone involved for the opportunity.

Please note that this interview has been edited for clarity and flow.

Thank you so much for this opportunity, Tanaka-san. Your recent market share data is quite impressive. What do you see as your widest user base, and what is your biggest target group?

Kenji Tanaka: Our biggest focus is to increase the [existing] market, not to segment it. The overall market is decreasing, and it’s been a challenge to increase the market, overall. We believe we can do so by targeting professionals and enthusiasts.

Amongst professionals and enthusiasts, who do you feel you still need to attract to your products?

KT: There are many types of professionals. Our target for the a7R II are landscape and portrait shooters. The a7S II has been adopted by many professional videographers. Right now we are already reaching these professionals successfully. But as of last year (2016), our products have gained significant capability in terms of speed. Therefore, we would like to target sports shooters and photojournalists by continually improving the speed, performance and usability of our products.

What about wedding and events shooters?

KT: Yes, these professionals have been adopting our products as well, and they’ve particularly appreciated Eye AF and the resolution of our products. However, they require more durability. We are actively working to address their needs.

Matt Parnell: In particular, one piece of feedback we frequently get from all of our wedding photographers is that the low light performance of our products has changed the way they can shoot events.

KT: And wedding videographers particularly in Asian countries have invested in our products.

Technologies like Eye AF depend upon close integration between the imaging sensor, processing pipeline and AF algorithms programmed into the camera. Sony’s close integration between the image sensor and camera divisions allow for quick iteration on such technologies. As photographers, we win: the compositional freedom Eye AF-C affords me for (unpredictable) newborn photography is unparalleled. Even with the pocketable RX100 V. Photo: Rishi Sanyal

Which camera do your users gravitate toward in particular for autofocus capability?

KT: Professionals largely use the a7R II for autofocus ability. a7S II for video.

The form factor of an a7S II and FS7 are very different. Do you see a convergence of these products and, if not, how will you target these user bases separately?

MP: We see many professionals and documentarians that use both the FS7 and a7S II to complement one another. A and B cameras, for example.

Daisuke Goh: But they are very different in terms of ergonomics and features, with Raw recording and high-frame rate being core technologies of the FS7. Those who need these gradually step up to the FS7, often from the a7S II.

You’ve stated your intention to re-stimulate this declining market via innovation. How do you think your competitors should do the same?

KT: I can’t comment on other companies, but I can explain our position. The most important thing is the image sensor. As you know well, every camera function is related to the image sensor. For example, the AF sensor and exposure sensor are all based off the image sensor. So the image sensor is key, and we develop it in-house. This means we can customize our future products with more intention [by having our camera and sensor development teams working together]. This is a differentiator compared to our competitors.

Speaking of sensors, are you interested in lowering ISO sensitivities? Particularly by increasing full-well capacities of pixels to increase dynamic range and achievable signal:noise ratios (a la Nikon D810)? I think many landscape and current medium format shooters would be interested in this.

KT: Both high and low ISO sensitivities are important. In case of low sensitivities, we are working on increasing saturation capacities, or well depth, of pixels. In case of high sensitivities, pixel size matters.

The future is certainly bright for Sony, and photography in general. One thing I’d love to see is extended (lower) base ISO dynamic range via higher saturation capacities of pixels, a la the Nikon D810 at ISO 64. While this a7R II certainly did the job perfectly well capturing this high dynamic range scene on my recent trip to Iceland, a lower base ISO that would have allowed me to give the sensor even more exposure before clipping the highlights to the left, yielding an even cleaner image. Technically speaking, an image with higher overall signal:noise ratio – the reason people love medium format images. It’s great to hear that Sony’s sensor division is fully aware of this need for lower, not just higher, sensitivities. Photo: Rishi Sanyal

But your a7R II, which has very similar low light stills performance to the a7S II, suggests otherwise. Could you elaborate?

KT: Think about 8K video. To get that sort of resolution on a sensor, you need larger sensors, otherwise pixel sizes are too small. To get 8K from a micro four thirds sensor, for example, the sizes of the pixels have to be very small, around 2.3 microns. I think larger sensors are important to maintain image quality as we go to higher resolution video and stills.

Right, it’s more sensor size that is key.

KT: Yes, this is why we choose to concentrate on full-frame.

Is 8K video something you’re already working on?

KT: We can’t comment on future product plans; however, we can confirm that we are paying close attention to all trends in the video marketplace, including 8K.

And you already have products that are sampling 6K: the a6500 oversamples its 4K footage by 2.4-fold yielding extremely crisp footage. Meanwhile we have some professional videographers intent on using high-end pro- 6K and 8K products to get oversampled 4K; yet you offer it in a consumer product. I find that interesting.

KT: Yes, we already have 6K sampling.

Are you also focusing on global shutter?

KT: Yes, that is one technique to remove rolling shutter artifacts. However, there are other choices to remove this artifact, like a mechanical shutter.

Increasingly we’re seeing computational approaches to get better imaging performance from limited hardware.¹ When do you feel computational technologies attempting to simulate the effects of larger sensors and optics will truly challenge more traditional approaches?

KT: These approaches work in some occasions, but it’s hard to realize for all scenarios. And a lot of the computational products I’ve seen so far need very large, fixed F-number primes [Editor’s note: presumably to compensate for smaller sensor sizes].

Are you trying to build in any computational approaches into your current cameras today?

KT: No. But speaking of light field, of course we are studying it. But not at a production level.

Do you think that as megapixel counts get higher and higher, it would make sense to devote some of those pixels to light field?

KT: There’s currently too much of a resolution cost. You need to devote at least a 5×5 pixel array just for one output pixel.

Do you benchmark against competitor products when developing your own products?

KT: Of course. We benchmark against the best product for any use-case. We learn from other companies. We must, because we are still beginners. The challengers.

Daisuke Goh, Manager, Product Planning & Business Management, Digital Imaging. Goh-san was product planner on the a7R II, and is pictured here looking out upon the Gulf of Thailand at moonrise during a recent trip to Sony’s Chonburi factory. To say the least, talking directly to these brilliant engineers was a unique opportunity to relay our, and our readers’, feedback for future iterations of Sony products. Photo: Rishi Sanyal

The convenience of the smartphone is a challenge to cameras. I’ll often see friends pick up their smartphone² instead of the ILC sitting in front of them to snap a photo of their child. What is Sony doing to help the parent, the hobbyist get assets off the camera and into a library accessible from all devices?

KT: We have apps like PlayMemories to make things easier for smartphone users. In the future, we cannot avoid making this process even easier via better integration with internet/cloud services. This is a big topic of discussion and something we are investing in. One issue with direct communication from camera to internet services is that regulation, not technical, issues make this difficult. Everyone already has a smartphone, so we want to use the smartphone.

One thing Sony can’t be blamed for is a lack of caring when it comes to quality. From Betamax to Blu-Ray to LCOS displays, how do you maintain a culture of insistence on quality and innovation across such a large company?

KT: For me, Sony’s founders are incredibly important. They are no longer alive, but the founders’ spirits and will are alive and well.

DG: Have you had a chance to read the founders’ spirits? It’s written. It basically says: ‘Always have a playful mind, and do something that others don’t do.’ This is basically in our blood.

That’s a great philosophy. Thank you for your time!

KT, DG, MP: Thank you!

Editor’s Note:

We nab every opportunity we can get to sit down with engineers at camera companies, and are particularly honored when we get a chance to speak with executives like Tanaka-san, who is head of Interchangeable lens (ILC) products globally, and Daisuke Goh, who was product planner on arguably one of the most exciting cameras we’ve seen in recent times: the a7R II.

Recent data over a two month span show Sony to have pulled ahead from #3 to #2 in full-frame ILC market share, based on revenue. Sony stresses this was no easy task, in particular given the shortages it recently experienced in delivering one of its most popular – and most revenue-generating thanks to its price – products due to the earthquake: the a7R II. It was interesting to hear Sony’s response to this (for them) exciting news: Tanaka-san stressed that the overall goal of Sony Digital Imaging products is to grow the (now declining) market, not segment it or pull ahead at the cost of its competitors. Sony believes it can do so through innovation, which should spur the expansion of the hobbyist segment. By offering imaging experiences and quality far above and beyond what is capable with smartphones, Sony hopes to rekindle interest in cameras and dedicated imaging products.

Given their focus on quality, it’s not surprising that Sony is pouring so much effort into their full frame products. It was only a little over 3 years ago that Sony launched the world’s first full frame mirrorless system, just a short year after launching the world’s first full frame camcorder (NEX-VG900). Since then, Sony has seen a remarkable adoption rate: they claim they are #1 in 4K camcorder sales, and hold over 80% of the mirrorless market share in North America. The largest adoption of full frame products has been in China and the US, and Sony’s projections estimate an additional growth of 30% in the full frame market. While some may argue that is optimistic, Sony cites the general market increase with the release of the Alpha 7R II, showing that innovation drives growth.

And Sony is particularly innovating in the image sensor sector, where they claim they are investing more than most. It’s interesting to note that the smartphone industry – the very one threatening dedicated imaging products – itself helps Sony, since Sony is a major supplier of smartphone camera sensors. So when it comes to image sensors, Sony’s return on investment is multi-fold: technologies like 3-layer stacked CMOS for smartphone cameras that allow 4x faster readout speeds than conventional chips for minimal rolling shutter and 1000 fps video capture will not only make our smartphones better, but will also trickle into ILC products and allow Sony to re-invest resulting earnings in even more exciting sensor technologies. It’s not just Sony that benefits from this – like smartphone manufacturers, other camera manufacturers also benefit from Sony sensor advances. What Sony has to offer though, as stressed by Tanaka-san, is the ability to work closely with the sensor team to develop better products and features around the strengths of those sensor developments. Autofocus and subject recognition improvements, for example.

Sony’s approach certainly appears sound: exciting technologies offering new imaging experiences spark the curiosity of not only enthusiasts and hobbyists, but professionals looking to differentiate their work as well. And many others as well: documenting the fleeting moments of our lives is arguably a very human interest. Devices that allow us to do so more easily, more readily, and in higher quality are certain to appeal even to the amateur mother or father capturing the irreplaceable moments of their little ones’ lives. With the iteration we’ve seen in Sony ILCs and premium compacts in just the last 3 years, it’s no surprise that Sony aims to be the #1 premium imaging company. And we will all benefit from its relentless drive.

Footnotes:

¹Lytro for example. More recent approaches include the Light L16, which combines lenses of multiple focal lengths to achieve high-res imagery and a large zoom range. The Google Pixel smartphone uses multi-imaging techniques to get impressive image quality out of a small sensor. iPhone 7 uses two lenses to create a depth map to simulate shallow depth-of-field. Computational approaches of recent are seriously challenging traditional cameras for general users that aren’t too too attentive to the outcome.

²Largely because of the ease of backup, curation, and sharing to services like Google/Apple Photos and Facebook.

Articles: Digital Photography Review (dpreview.com)