Roger's Corner

Is Sony Going to be the Digital Kodak?

Published July 11, 2012

My mind is a strange and dangerous place. I shouldn’t go in there alone after dark. But the other night I was thinking, just me and myself, about all the new camera releases this year.

Which has made the biggest impact? Is it the Canon 5D III with that world-class autofocus system? The Olympus OM-D that has brought mirrorless cameras up a notch in image quality and usability? Should I mention the excellent Samsung NX20 just because no one knows it’s really good? Should I give the Fuji X-Pro an award for best concept most poorly carried out? Consider the Sony NEX-7 for putting full-frame resolution on an APS-C sensor?

So, yeah, I was going to give all those cameras a fair trial and then, of course, decide that the Nikon D800E would win with its one-two punch of highest resolution 35mm sensor ever and gutsy reduced-AA-filter decision. After all, you know me. I’m all about resolution. Even more than just resolution, though, every analysis of the D800E shows amazing dynamic range, low noise, you name it (1, 2). No question, it’s a landmark camera.

But then I got to thinking. (I hate when that happens, but sometimes I can’t help it.) Isn’t the D800E mostly about the sensor? So is it Nikon or Sony (or both) that deserves the credit?

We’ll pause here for a second to let the Nikon Fanboys gear up and send for some villagers with torches and pitchforks.

While we’re waiting, perhaps you’d like to use the internet and search for how many CMOS patents Nikon has registered during the last few years.

Nikon makes and designs some of their CMOS sensors and all of their signal-processing and image-processing chips. But Nikon does not have many CMOS sensor patents. Sony has lots. In fact, they’ve registered 64 CMOS patents (3), just in the last year—far more than any other company. And the Exmor patents that seem to be the driving force behind a number of cameras are all Sony’s.

This isn’t bashing Nikon. They make a great camera and they’re smart enough to put a great sensor in it. I doubt they designed the cloth the camera strap is made out of either. They chose it because they thought it was the best cloth; they didn’t have to open up a weaving plant to get it. They chose the sensor in the D800Es the same way, by designing the processing chips and circuitry around it, then the camera that houses it.

It turns out that another of my “great change” cameras has a Sony sensor inside: the Olympus OM-D.

The OM-D sensor seems to have better dynamic range and signal-to-noise ratio than the other micro 4/3 cameras (45), which apparently all have used Panasonic sensors until now.

The newer Pentax K cameras I’ve read rave reviews about (I haven’t had the pleasure of shooting one myself) have Sony sensors in them apparently, instead of the Samsung sensors used in previous Pentax cameras. And I guess it’s obvious the NEX-7 uses a Sony sensor.

What do you mean by digital Kodak?

I mean this in a positive way. We tend lately to think of companies “going Kodak” as a bad thing: riding a cash cow into bankruptcy while short-sighted managers ignore new technology that’s taking away your market.

We forget, though, that for nearly three generations Kodak was the photography company. They were cutting-edge. First developing dry plates and then film, Kodak released “consumer” cameras that brought photography to the masses and made everyone’s prints. They dominated the photographic and movie industries for nearly 100 years.

They were a camera manufacturer, of course, but that was really just a side business. Even if your camera wasn’t a Kodak, chances were that you made your images on Kodak film and developed them using Kodak processes.

Kodak made all of the prints, first using their paper in enlargers then eventually in automated machines in camera shops and drug stores. Whoever sold you a camera made a bit of money, of course. One time. Then you paid Kodak over and over again, as they sold you film and made you prints.

It was a great business model, and Kodak was printing money. Literally.

During the Great Depression Kodak bankrolled the city of Rochester, N.Y., allowing it to issue its own “scrip” that could be spent just like dollars anywhere in the city. It was said the depression never came to Rochester, largely because Kodak continued to grow and do well during that time.



Many older film photographers remember Fujifilm as a viable Kodak competitor. They were, starting in the 1980s, but Kodak had already been around 100 years by then.

Up until that time Kodak sold 90 percent of the world’s film (6). Few people today really appreciate how huge Kodak was (partly because Kodak was rather secretive about how huge they were, back in the day).

At one time Kodak employed 60,000 people—just in Rochester, N.Y. (7).  To put that in some perspective, FedEx is by far the largest employer in my hometown of Memphis, Tenn., where they employ 30,000 people.

OK, we get the film-sensor analogy. It’s kind of weak.

Yeah, I guess it is. Sensors are processed by chips in the camera or programs on your computer. Sometimes they are printed.

It’s not like the days when every roll of Kodak film got sent through baths of Kodak developer to be printed on Kodak machines. (Man, was that a great business or what? No wonder they got a bit arrogant.)

But like the days when lots of different cameras had Kodak film inside, lots of different cameras today have Sony’s CMOS chips inside (like higher-end Nikon cameras, the Olympus OM-D, and, of course, Sony’s cameras.) Even the cameras in iPhones (8) and many other cell phones have Sony CMOS chips.

Oh, and not to mention the Sony sensors in a number of point-and-shoot cameras, Sony and otherwise. Not to mention video cameras. Or security cameras. You get the idea.

In 2010, Sony accounted for 33 percent of the world’s imaging-sensor sales (both CMOS and CCD). In 2011, it was 37.3 percent. Canon is a distant second with 12 percent. Samsung, at 11 percent is the only other manufacturer with over 10 percent of the market (9).

And Sony has just invested a billion dollars to increase capacity in their Nagasaki plant to 60,000 wafers a month (10).

It’s not just quantity; it also appears that Sony has a quality leap over most other CMOS manufacturers. FWIW, DxO Mark shows 13 cameras brought out in the last two years that have Dynamic Range of more than 13 EVs.

Four are Sony cameras, two are Pentax cameras using Sony sensors, and six are Nikon cameras. Of the six Nikons, four (D800, D800E, D7000 and D5100) are known to use Sony sensors. The other two are the D4 and D3200, which I’m not certain about.

Maybe a better analogy would have been the old “Intel Inside” campaigns in PCs back in the day. If Sony keeps it up, we may see camera brands bragging about having a Sony sensor inside, instead of just letting it slip out in a meeting or press conference or having websites X-ray the sensors to find the brand (11).

Or maybe not. It’s the camera business after all, and today’s unsurpassable technology is tomorrow’s also-ran.

So what’s the significance?

Heck, I don’t know. I do know competition is a good thing. And if we have lots of camera manufacturers able to get excellent imaging sensors to design their cameras around, that’s got to be good.

Plus I’m sure that Canon, Samsung, Panasonic and the other imaging chipmakers are paying attention and trying to stay ahead of the curve.

I’m not too concerned that Sony will suddenly cut other camera manufacturers off from their chips; they’ve obviously invested a lot of money in making them so they need to keep selling them.

<Begin humor>

Maybe they’ll cut some deals, like “we’ll supply the chips for your new camera , if you add our new memory card format to it.” Something like that.

What would be very cool, I think, is a deal like “we’ll supply the chips for your camera, if you supply us some good lens designs for our NEX cameras.” It would be a win-win.

Like one of my bosses told me when I told him I was quitting to take another job: “I think it’s great that you’re leaving us to go work for them. It will improve the average IQ in both places.”



Roger Cicala

July, 2012

Author: Roger Cicala

I’m Roger and I am the founder of Hailed as one of the optic nerds here, I enjoy shooting collimated light through 30X microscope objectives in my spare time. When I do take real pictures I like using something different: a Medium format, or Pentax K1, or a Sony RX1R.

Posted in Roger's Corner
  • Rasta

    Hello,first I believe canon “hand” was forced when they created the 5D mk3 and placing the AF in that model, however I feel they already had the AF especially if we look at there 1D models even in the professional film bodies they have out there(sidebar; I like the eye-control AF in the 35mm). I believe canon is very aggressive when it comes to placing new products in the markets. I will say I did not know Nikon was obtaining their sensor from or possibly from Sony, which would mean Nikon only has to worry about making it work in their system. Which I say,”half the battle” is done cause the sensor is outsourced. Whereas Canon has to make the developmental progresses first then integrate. Saying all that to say,it’s very likely Sony could gain such dominance in the digital world. Lastly,I feel Canon got stuff up their sleeves still…. Like the rumors of 75mp camera,but the funny thing they had 100+mp camera prototype on deck video like a year or so ago(I should of saved it,darn it). Please excuse some of the typos

  • To follow up on what Mike S said: a lot of the large companies with huge patent holding have patent cross-licensing agreements

    They do this so they can avoid the sort of problem that you see with patent disputes. The larger your patent holding (in the same field as another company B) the more likely it is if company A accuses company B of patent infringment that company B will search through their patent portfolio and find a interpretation of a company B patent that company A is infringing. Either at that point they decide to settle and cross-license or they go to court (usually if there are other strategic factors involved … see Apple vs Samsung).

    For Sony and Nikon there are lots of patents and lots of fields to collide in: lens design; sensor design and technology; AF tech; VR/OSS tech, etc.

    The step up from meere cross-licensing is an OEM relationship where you buy product from another company (using their tech) designed to your specs.

    The next step is technology cooperation (which goes both ways) to develop tech that’s of interest to both companies with each contrinuting particular tech to the design which then may be more widley licensed or just used by the two companies involved.

    Nikon clearly is working with Aptina too (Nikon 1) and is current shipping the D7000 with a Sony sensor and the P7100 (and probably the upcoming P7200 too). Some of the Nikon designed sensors may be a mix of Nikon tech (the basic design approach) and particular licensed Sony and Aptina tech. Without Nikon telling us we can’t know.

    Another thing to keep in mind is dyes used in the Bayer filters are different across difference sensors and different manufacturers. You could take a straight OEM Sony sensor design and ask for your own (patented) dyes to be used in the filters. A perfect mix of tech.

    But one thing seems clear: Sony are building the sensors camera designers want.

  • Frozen I

    I see someone had the typo “censor” for “sensor” – I find that amusing because “censor” has two meanings: the person who decides what you are allowed to see, and an incense holder/burner (often used in religious ceremonies. Makes me imagine an element in my digital camera which is blocking me from shooting certain images, or burning incense in a prayer to “get the shot” 🙂

    Yes, I’m thinking too hard about a single letter typo…

    I don’t regret buying a D800E, although I would have loved an option to get the same sensor in a Canon 1 series body (I have a lot of Canon glass already). Unlikely that I’ll ever get the chance. Not like being able to choose whose film went into my SLR…

  • Mike S.

    I think something that has to be considered is that Nikon Precision (the division that makes the semiconductor stepper equipment that makes sensors) has a very big customer: Sony. And Nikon is one of Sony semi-conductor divisions biggest customers. There is, from what I hear, quite a bit of technology exchange between the two; something that isn’t very common in the USA, but happens more frequently in Japan. Thus, I think the success of the D800 is not just Sony; it’s a product of what has worked out to be a very successful relationship between the two companies.

  • TomH

    I also hope other manufacturers take note of the low noise of Sony sensors, especially at low ISOs. I have an Olympus SLR that is about 7 years old now, and the Kodak CCD in it has visibly lower noise than almost all new cameras, even those with larger sensors. It’s terrible above ISO 100, but that low noise at low ISO is what I need for the photography I do.

    The reason I have not upgraded in all that time is because I’m waiting for a camera with 2 main things: A sensor with low noise at base ISO, and EFCS (electronic first curtain shutter). Both are needed for macro photo stacking because any noise is greatly amplified (even with filtering), and without EFCS the shutter mechanism bounces the camera so much at higher magnifications that there’s no way to get a sharp image. This is frustrating because Canon only use their own sensors which have too much patterned noise at base ISO, but have EFCS, and Nikon because they use the great Sony sensors, but do not have EFCS.

    However, it looks like the Sony nex-5n has both a good sensor and EFCS. That’s why I’m looking into buying one: Because they put these 2 features that make better pictures into the same camera, which is more than anyone else has done in the better part of a decade.

  • Peter Dove

    You missed a direct analogy between Kodak and Sony: Sony’s recent sensors are lauded for their low noise, even at low ISO, while Kodak was (as far as I know) the leading manufacturer of … low-noise sensors. Their KAF line has been used extensively in research-grade astronomical and histological cameras for years, with Sony and a European company I can’t recall right now also grabbing some of the action. We’ll see how well their Truesense spinoff ( can carry on.

  • Jack Stivers

    Please, please Sony, do trade sensors to Nikon in exchange for e-mount glass. Sony releases “e” lens at a glacial pace.

  • TRV

    The problem for Sony is that unlike for Kodak and film, the sensor is not the high value-added part of the camera sale and, as Roger correctly points out, there is no ongoing transaction with the end customer. This wouldn’t matter if Sony cameras were also market-share leaders, but they’re not (being a distant third to Canon and Nikon).

    Whilst Sony evidently feel they are onto a good thing with their SLT and NEX cameras, I’ve yet to be persuaded that they have a “killer app” that will win buyers over. This could be the Exmor sensors, but their semiconductors division seems happy to sell the “crown jewels” on to the highest bidder. Sony is currently struggling for direction and not in the best financial state. Whilst I am sure that they appreciate the income from sensor sales, I would implement a policy of “if you want a Sony sensor, you need a Sony camera” (remember the iPhone commercials on TV?). It’s unlikely that Sony will cut-off third-party manufacturers, as this is not the Japanese business culture.

    Sony need to succeed in imaging, as they have stated that they are staking a large part of their future strategy upon growth in this area. Many people think that Kodak failed because they stuck their heads in the sand with regard to the digital revolution and failed to exploit their early lead. This is partly true, but selling digital sensors would never have been a big enough market to support a company the size of Kodak.

    I think that Sony need to think carefully about their vertical position in the market, they need to be market leaders in cameras and camera phones, not just sensors. This means recreating the premium brand image that Sony used to enjoy in consumer electronics, but has now lost to Apple and others. I believe that taking back proprietary technology into their own product lines would help achieve this.

  • Esa Tuunanen

    XQD is huge amount better memory card format than completely overhyped Shit Digital.
    Being based to well documented computer data bus gives it proper flexibility for future. Already 2010 finalized PCI-E v3.0 upped data rate to gigabyte per second and devices/hosts are backwards compatible.
    And storage capacity limit would be same as for computer storage devices. And we all know how computer HDDs got limited to 128GB… Which was actually limit in first 17 years old CF specification which has since then been flexibly extended along with Parallal-ATA standard.

    For comparison every new specification of MMC/SD has broken compatibility with older hosts.
    And physically flimsy SD is inconveniently small to handle (don’t even want to think about handling it outside at winter) and small physical volume also prevents putting more memory chips in “RAID 0” for lot faster read/write performance and capacity with same memory tech.

    So I’m sure hoping Sony or someone else pushes XQD down the throats of average consumer dummies who don’t know a shit even if they fall on their faces into it when marketroids just tell them that it’s the latest fashion trend food!

  • Dave

    Another Rochester native here. Best quip I heard about Kodak is, “they make their money on your BAD photos.” Meaning, every time you click the shutter, they get paid, regardless of whether it’s a good photo or not. Ponder that with our modern DSLR mindsets – after initial investments, every click is free to us! (except for maybe wearing out the shutter or needing more hard drive space).

  • Ken Owen

    At least Memory Stick wasn’t rammed down everyone’s throat. However, is Sony now in a better position to XQD us all?

  • George Lien

    Great Analogy and good insight !

    Thank you for a good read.


    George Lien

  • At first, Please forgive me for my poor English 🙂
    As I’m keen on progress I suggest to read blog
    As I’m keen on precision I read your blog about lenses testing and variation.

    So this is my thoughts about sensors

    preparing a new sensor is very cost full process as I got information its can go up to $50M, and it’s only a part of making new camera. So outsourcing those cost is good idea. But then you could lost secret know how and advantage of competitors.

    Let’s look at Sony APSC 16MP sensor case-study:
    Almost the same sensor in Nikon D5100 D7000 Pentax K-5, K-01, K-30, Sony Nex5 series, Sony SLT A66 ect. It’s a huge production volume. So cost of development by for one unit will me smaller.

    And they are not the same. You will get different image form Sony from Nikon from Pentax. So still many thinks relay on those little-big things surrounding the sensor.

    Main question is: how much those cameras are working in the same area of market?
    In my opion: they almost not interfere. So Sony make they case-study.
    I thing we will see more cameras with sensor from OMD-5, from D800 also.
    Personally I’m waiting for FF from Pentax. I hope to see there crop of sensor from 645D made by Kodak. Sony have still work to do with red color :).
    Personally I’m holding my thumbs for Canon, Aptina, Panasonic,that they make new innovation to get the level of Sony.
    But for me Kodak still have very strong sensors for still image.

  • “My mind is a strange and dangerous place” Look on the bright side: the first step to good mental health is realising that you have a problem 🙂

  • Markus

    Sadly enough – Kodak was (is?) even making sensors (CCD’s, e.g. for MF & Leica), too…

  • Ed

    Have the villagers with torches and pitchforks arrived already?

  • A

    It’ll be interesting to see how the cameraphone market matures.

    The cameraphones are slowly eating market share from the compact cameras, and the most interesting sensor there is by Toshiba. That’s the sensor behind the Nokia 808 PureView.

    That said, Sony (and Samsung) manufacture cameraphone sensors too.

    I suspect the loser may be Canon – they have no cameraphone sensors at all.

  • Carsten

    Fuji’s XTrans sensor in their X-Pro1 is afaik also nothing more than the usual 16 MPix Sony sensor, just with a different color array in front of it. Eyes from Sony, glasses from Fuji. 😉

  • Esa Tuunanen

    Competitive sensor is indeed one of the most important pieces of high end digicams.
    If Olympus had had equally up to date sensor five years ago 4/3 DSLRs would have no doubt fared better.

    Other manafacturers definitely have to keep improving their sensors if they want to remain competitive.
    Samsung has huge resources and also Panasonic is actually electronics giant fully comparable to Sony so it isn’t about their research capability being out of league.
    Todays very strong IP/patent laws might be bigger obstacles and preventing them from using results of their research simply because Sony got there first.
    So might well require advance of manufacturing technology making possible notably different sensor before competition has change to rise past Sony.

  • The big question is, will sony maintain their front-runner status in censor market. There was a time not long ago that canon arguably made the best CMOS sensors for DSLR cameras and they still manage to do really well on the high ISO imaging. However, you wonder if Canon’s success made them lethargic and less risk-averse. I recall attending one of these big tech shows (can’t mention the name) where canon had a huge presence. This was around 2004-2005. I asked them if they would ever put a sensor like their 20D into a smaller P&S camera. The canon rep said, they get that request a lot but Japan seem to ignore it. One reason being the price. He then asked me how much I would be willing to pay for it. I replied $800-1000 for a camera like that.

    Fastforward and it took canon like 8 years to release that type of camera after the market began saturated with competing products. Basically playing a copier than innovators. This the problem with most front-running companies. They become risk-aversed for various reasons. Fear of internal cannibalization, to hubris to internal politics, etc. So we have to see how Sony would react to being a front-runner while remembering there was a time when Kodak made the best films in the market.

  • BetaMax was actually incredibly successful. In the professional broadcast world.

    Ultimately, Sony made a whole lot more money from BetaMax than JVC did from VHS.

  • Roger Cicala

    Chris, those are interesting thoughts. It looks like it could be that way. I know Canon has a large market share and I assume (but don’t know) that’s just from their own cameras. But they might be making some cell-phone or other companies point and shoots, too. Sony seems to have it going on right now, but we all know how long that lasts in the camera business. You’ve got the lead until the next product cycle, then who knows?

  • intrnst

    It seems that Sony actually learned from the historic Betamax failure (and Xerox’s and IBM’s…): Adapt or die.
    Now is divide and conquer, infiltrate and invade… or just smart symbiotic strategy? (3S! Uh, nice)

    Time — the next leap forward — will tell.

    Nice reading, as always.

    “It is not the strongest species that survive, nor the most intelligent, but the ones most responsive to change”
    Charles Robert Darwin

  • Great comments Roger. As someone who went to the U of Rochester, I can appreciate how it was essentially Kodak City (the pollution…). The city of Rochester is in a sad state especially because of the general decline of the industrial north. Of the big three, Xerox, Kodak, and Bausch & Lomb, only the last one has a significant presence any more.

    I think Sony is more like Intel, they’re just outperforming and outpricing everyone because of their significant investment.

    Just like Kodak though, there will probably some sort of disruptive technology that gets invented in the next 10-20 years that might make their lead obsolete.

  • Chris K

    My first thought was: Canon doesn’t want to play in this market. If the proliferation of Sony sensors comes to pass, we’ll see:

    Canon sensors in Canon cameras
    Panasonic sensors in Panasonic cameras
    Sony sensors in everybody else’s cameras

    Which company do you think will produce the best sensors long-term? And which company do you think will have the most leverage over the camera industry?

    Is Canon being wise by allowing their excellent sensors only in their own cameras? Or are they missing out on a ton of money that could be put towards sensor R&D, which will ultimately push the competition ahead of them?

    I must say, I’ve owned (Micro) Four Thirds cameras from the E-410 all the way up to the EM5. The EM5 is the first Four Thirds-size sensor I’ve been truly happy with when compared to its Canon contemporaries. (EM5 vs 7D and 5D2 in my household.) It feels almost like parity, which is extraordinary when you place the cameras side-by-side.

Follow on Feedly