Nikon now producing their own sensors?

The new 14.2MP DX-format CMOS sensor in the Nikon D3100 could be produced by Nikon

Thom Hogan has an update on his Nikon D3100 post indicating that based on one of his sources, Nikon has indeed started to produce their own sensors. I already mentioned a quote from Nikon's website that the new 14.2MP DX-format CMOS image sensor from the Nikon D3100 is "developed by Nikon", although from the press release it is not clear if it was also produced by Nikon. The whole sensor fiasco started  few weeks ago when rumors started to float around that Sony will stop producing full frame sensors, which could put Nikon in a difficult position.

One of the comments from the Q&A session in Denmark regarding the sensor in the D3100 was "I think that it is a Sony sensor".

Sony is expecting to announce new cameras next week with 14MP and 16MP sensors inside and same ISO range as the D3100 (100-12800). The D90 replacement is expected to have a 16MP sensor. Coincidence? I don't know.

FYI: when the Nikon D3x was released back in 2008, Nikon made a statement to Rob Galbraith and CNET that the D3x sensor was "..with final production executed by Sony". As fa as I know, those are the only official statements from Nikon indicating that Sony has any involvement with their sensors.

Update: here is one more link that I missed where Nikon officially mentioned that they are trying to push their own sensors:

"We have a longstanding relationship with Sony. If the sensors for Nikon D3s, D3 and D700 are designed by Nikon, Nikon D3x and those of the small APS-C sensors are from Sony. We want to use our own sensors in SLRs most popular [small sensor APS-C, Ed], as the performance of our sensors are better. However, it will take some time as it takes to achieve economies of scale."

This entry was posted in Nikon D3100. Bookmark the permalink. Trackbacks are closed, but you can post a comment.
  • Vladi

    Possible. But now rumors circulating of new 16mpx FX sensor that should be found in D800.
    Sony doesn’t have this kind of sensor, so maybe new Nikon sensor?

    • Denko

      Looking at the D3100 sensor and especially the glass ontop of the array it is clear that the entire assembly would work for an FF sensor too (same outer components.) This to me is evidence that Nikon are smart enough to do things cost effective and by doing it that way it is VERY possible that an FF sensor will be just slightly more expensive to make (pending any yield issues… Nikon is in full control of the lithography so if anyone can do it it’s Nikon, from this yield perspective.) I definitely will welcome FX to the masses… I have felt slightly neutered using DX because I grew up on film… yeah obviously I don’t care about reach and the necessity to use VR on the fringes.

      Note also the “Steady shot inside” on the new Sony’s … clearly optimized for video. Nikon would never do that and them going at it their way is the ONLY sensible thing to do. The wait is a long one for me… but I will be waiting a bit longer because I know Nikon does have something that will cause a stir.

      As I see the “future”…
      Sony optimized on video and will kick butt in that arena… will wipe Canons “lead” in that respect… the image quality will however be very lacking.

      Nikon will be diametrically the opposite to Sony and will have their cameras optimized for single image perfection and will further go away from Canon’s pictures.

      Canon will loose unless they drastically fix their line of diverse censors that quite frankly are all over the place… they will need to trim some fat… actually a lot of it. 😀

      Anyhow not an analyst, just a dude that get occasional epiphanies.

      • The Man from Mandrem


        You epiphany is flawed. Nikon’s presence in litho does not translate to a tangible benefit in manufacturing image CMOS. Semiconductor capital equipment is extremely competitive and the bigger parts of the market are Memory and Logic. If Nikon were stupid enough to hold back Litho innovation to give themselves a leg up in making Image CMOS, while losing market postion in the Litho business, they’d put themselves out of that business. You seem to have no idea how expensive Litho tools are. Good try though.

        • My guess is that we’ll see a Nikon camcorder with an F-mount soon. Similar to the thingy Sony announced.

      • elliot

        “Note also the “Steady shot inside” on the new Sony’s … clearly optimized for video.”

        All Sony’s DSLRs have SteadyShot Inside (previously called Super SteadyShot) … and they always did.

        As for optimizing video, what DLSR manufacturer isn’t these days?

        • ^Nikon. haha – And we love them for focusing on what they do best!

          • Johan Krüger-Haglert

            “^Nikon. haha – And we love them for focusing on what they do best!”

            … lag behind?

      • BornOptimist

        No it’s not big enough for a FF sensor.
        Just measure on your screen. I used a ruler on my screen, and the sensor width is 70mm (on the screen). This is a 24.1mm wide sensor (approx 3mm on screen equals 1mm sensor). A 36mm wide sensor will therefore be 109mm wide (that would be the width of the sensor area on this magnification).
        The whole black area on the sensor module is 110mm wide, so there are not nearly enough place for a FF sensor on this module. The pins in the sensor to attach it to the module would make it much much wider than 110mm

        • Denko

          Sadly you are right… smallest crop factor would be 1.35… to fit it in the same assembly. ~17.5MP with same pitch… skrewing up all lenses if they did… so… no… this pie in the sky seems this morning improbable to happen. Ah well it was a good night sleep 😀

  • NiknWontRepairMyGray

    So we should expect two more cameras coming?

    Both are expected to be DSLR or will one of them be an EVIL?

    • NiknWontRepairMyGray

      Nevermind, the article said 2 Sony will be announced. This D95 anticipation is driving me crazy.

  • Stinger

    was expecting to see the same sensor from NEX-5 in Nikon D3100 but it looks like it is slightly different.

    23.4×15.6mm (NEX-5) vs 23.1×15.4mm (D3100).

    Can it be the same sensor and if it is so why the size is different?

    • Ant

      If it’s a different size then the design is fundamentally different, unless there’s dead space on the NEX-5 sensor, which is highly unlikely, because non-utilized space just reduces yield.

      • PHB

        People are so confused about the Fab thing.

        It has the Sony name on the door, but it will be a totally separate business that has its own shareholders. Sony concerns will have a majority shareholding of course, but I would expect that Nikon also holds a stake.

        The way a fab works is that you bring your design in Cadence or whatever and they make up masks and make your chips in whatever volume you asked for. Using the Sony fab does not mean that you are going to get the rights to use the Sony chip designs – not unless you want to buy their standard cells (components) or use of their masks. And Sony won’t get to use the Nikon design either.

        When Nikon first switched to using the Sony design, they did base their chips on Sony chips that were known to work with that fab. But they were both heavily modified by the time of the D3/D300 launch.

    • These size discrepancies could all be BS. The D3x and A900/A850 sensors quote different megapixel sizes, but they are obviously related and output the exact same pixel dimensions (although they are decidedly NOT the identical).

      • BornOptimist

        All sensors has pixels around the edge that is behind a black layer. These pixels are used for noise reduction among other thing (it was described in a Nikon patent I read some time ago). How many that is used for this varies from sensor to sensor. They are still a part of the sensor, so if you compare the size of A900/A850 and D3X, you see they are the same. The difference is how many pixels are behind a dark fram. This dark fram is not a part of the underlying sensor and is different on the Sony-version than the Nikon version. The sensor can still be the same.

    • Note that the video specs are different, too. This to me is a more important distinction in terms of design than pixels, as photosites and their underlying sensels are easily scaled. Direct access to the photosite data, which is needed for video, indicates different electronics at the photosite.

      • Sahaja

        At the time the NEX 3 & 5 came out somebody mentioned that Sony had a newer and better APS sensor than the one used in those cameras and we’d see this in cameras they released later in the year – possiblyin a NEX 7. I guess time will tell – but could it be the D3100 sensor is based on a newer Sony sensor than the one used in the NEX 3 & 5?

        Does it matter who designs an makes the sensor? So long as Nikon uses the best sensor available (with or without their own tweaks) for the price and market segment a particular camera is aimed at – do most customers care?

        I have an Apple laptop and an HP laptop both using the same Intel processor (and a lot of other near-identical components) ~ yet they are quite different because of the software (operating system) and other design choices in each system.

        Just as Intel now have a virtual monopoly on the market for PC CPUs – could we one day end up with sensors in all cameras being made by just one or two manufacturers? I’m sure their are several manufactrers of imaging sensors who would love to become the Intel of digital photography. In the long term there might be a lot more money in that than there is in making cameras.

        Nikon developing all of its own imaging sensors could be a risky
        business – that sort of development costs a lot of money and some things they might want to do must already be patented by their rivals – and these things could be expensive to license or difficult to work around.

        If Nikon ploughs a lot of money into sensor development and sombody comes out with something substantially better – perhaps through some unexpected inovation – they could take a big hit.

        Could end up being a bit like betting the farm…

        • Yeah right, it doesn’t matter where the sensor comes from – as long as it the best solution for the money. Should Nikon even focus on manufacturing their own sensors? Well, as long as DSLRs do not feature interchangeable sensors maybe. A photographer should not have to buy a new camera just because he/she wants the more expensive film/sensor. The main reason why we are still in the stone age of digital photography are probably the Japanese. Open standards or an open interface are not an option, because it would be admitting that others can do the job just as well or even better. Doing something a bit different is not an option, because it would be a disgrace to the elderly engineers. Doing what the customers want is not an option because one has to be as stubborn as a mule.
          Maybe it is about time to start something like RED did in the film industry – to reinvent the DSLR industry. Call it PINK or something. I don’t know.

          • The Man from Mandrem

            Wonderful critique of Apple. Never realized they were Japanese. I’ll bring my passport next time I drive through Cupertino.

          • Jabs

            @David Hasselblaff.
            There is a simple flaw in your argument.
            If you have interchangeable sensors, then there is an ALIGNMENT issue with each change of sensor in the SAME camera.
            Optical dimensions, size, distances, etc., are fixed or we would not have interchangeable lens that work.
            Interchangeable sensors imply that the Technology has become a ‘commodity’ and no longer important, so many make the Mac. -vs- PC mindset argument via a shared CPU or ‘X86 instruction brain’.
            The problem with MOST arguments here is that almost everyone thinks in TERMS of desktop or laptop computers.
            Sensors are NOT CPU’s.
            Sensors are LIGHT gatherers and often AMPLIFIERS – not CPU’s!
            Sensors CONNECT to specific ‘CPU’s’ (not computer CPU’s such as those made by ARM, Intel or AMD) in a camera. These CPU’s might be proprietary and thus the idea of an interchangeable unit implies standardized components built according to Industry Groups, allowing MIX and MATCH.

      • Jabs

        Hey Thom,
        Perhaps the best analogy that I can make is in HOW AMD split up into a Chip Design House and a Semi Conductor or Chip making Company.
        AMD and Global Foundry = Nikon and Sony.
        One DESIGNS and the other produces the designs. They perhaps use the same Foundry for Production but Independent designs, though the Production TOOLS (software, etc.,), Equipment and such are similar for them to have a coherent WORKFLOW or Process Flow.
        Independent, yet Inter-dependent.
        Now, Global Foundry is majority owned by the UAE Government and supplies the CAPITAL spending needed in today’s stratospheric semiconductor world and AMD provides the design EXPERTISE.

        Nikon = Design House
        Sony = Foundry and Chip MAKING expert.
        Nikon = Camera expert
        Sony = Video expert

        PERFECT match, perhaps!

        • No, it’s more analogous to the AT&T Microelectronics Hobbit processor. When AT&T corporate shut down Penpoint, a primary consumer of Hobbits (via current and future Eo devices, including a <1 lb cellphone computer), AT&T Micro lost the prime customer and thus the economics fell in the tank and they (mostly) ended Hobbit R&D.

          One of Sony's problems (and many large companies have this issue if they create both parts and finished devices) is that can't count on themselves as the only consumer of the parts. Doing so would force a massive write down of assets, I think (or course, so might losing enough customers for their parts ;~).

          • Jabs

            @Thom Hogan.
            Both of us have highlighted similarities in those breakups but we are talking six of one or half a dozen of the other.
            Global Foundries is now making chips for competitors of AMD and has now become a full-time almost Independent CONTRACT Fab, just like Sony seems to be moving towards, hence that was my POINT!

          • Sahaja

            Hmm. With sensors like this possible, maybe Sony are indeed planning on getting out of FX (and maybe smart to do so):

            Can any lens resolve that kind of detail? If so FX could become the reserve of those currently shooting with medium format backs within a few years.

  • Anonymous

    Hmm, interesting, very interesting…

  • ha

    the rummor about sony stopping full frame sensor is just wishful thinking from “the other” guys who are scared of a nikon-sony partnership. sony is off course a much bigger company than the “other” guys so it is understandable why they fear it given sony competes in the pro video segment and has a much larger follower base than “the other” guys.

    • Roger

      Tell me about it. I’ll be laughing my ass off when I see the next Sony full frame. “Sony is dropping full frame sensors!”, lol. Yeah, right! Some people will publish any garbage as news, no matter how illogical it is (hear this, Thom or Lloyd?)

      • Rosco

        “I’ll be laughing my ass off when I see the next Sony full frame. “

        If a new FF camera comes out it doesn’t mean they haven’t dropped the idea of full frame sensors. It probably means that the camera’s been in development for the last 1-2 years.

        A bit like some saying the release of new Sony FF lenses means the rumour is false. Those lenses (and probably more to come) have probably been in development for a year or two.

        I’m not saying that they are halting making FF sensors, because I wouldn’t know what they are doing. Just that new FF cameras and lenses doesn’t prove the continuity of FF sensors. Cheers 🙂

        • PHB

          I don’t think it makes sense for Sony to make FF from a business point of view. They are not a top tier manufacturer in the DSLR space and will take a minimum of another ten years plus hope for either Nikon or Canon to mess up in a very big way.

          By 2020, the DSLR space will probably not be a consumer space at all. I expect EVIL type cameras to dominate the consumer space within 5 years. Those are the cameras that will sell in the millions. It will simply not make sense for Sony to compete in that space. They are a consumer brand, not a professional brand.

          I think we are going to see quite a few brands drop out of the DSLR space in the next few years. Not Canon or Nikon of course, but Pentax and Panasonic could go. And things are likely to be rather more Darwinian in the medium format world.

          Whether Sony drops FF has absolutely no bearing on what Nikon does. The Fab makes chips of all shapes and sizes and will continue to do so. Size has an impact on yield and thus cost, but from a technical point of view its just a setting on the machine that scores the chips on the back and cuts them up.

          • Roger

            For sure.

            I’d expect EVIL cameras to replace the “high end” compacts (I use the “high end” term loosely, as all compacts are crap) and also the low end DSLRs. I’m thinking, average person/soccer mom who buys low end DSLR cameras couldnt care less about quality, it’ll be easy for manufacturers to sell them EVIL cameras with those horrendous electronic finders.

            Medium format? I’m amazed it still exists, considering the (lack of) quality and prices. Anything other than Pentax should disappear fairly quickly. But rich medium format users are a strange crowd, they’re just a tiny bit insane. They went through all the Leicas, Canons and Nikons, so now it’s medium format they’re using and they’ve convinced themselves their Phase One or Hasselblad has like 6 stops more DR than D3x (LOL). Medium format companies live on this kind of stupidity and lack of objective comparisons, their existence literally depends on it.

            • Kingyo

              try 12.5 stops DR!.. have you even shot with medium format digital phase one backs? I doubt it. I’ve used the P40+ back and can assure you that the image quality & dynamic range are far superior than either the D3x or equivalent Canon dslrs. Actually the quality is ridiculously good 🙂
              It’s pointless to compare medium format digital to dslrs because their applications/advantages of use are different. I wouldn’t use an H2 or Mamiya 645df for sports or night photography any more than I’d use a D3s for advertising shots or high-end fashion. Medium format does quite well. You don’t hear anyone complaining that their 65 megapixel phase one back is lacking in pixel count. Their market can afford to buy or rent the equipment.
              For the rest of us that need speed, versatility, and high ISO + industry acceptable IQ, the dslr is perfect.
              I agree that Nikon and Canon will hold strong in the slr market, but I miss the days when both sides tried to out-do each other. This would have already given us the camera we’re looking for: high mp, high IQ, high ISO performance, and high quality video. (and a year ago I was the one saying “I don’t need video in an slr, if you want good video buy a video camera!”…that’s until I used the T1i/T2i 😉 which is why I’m hoping for a worthy FX w/ at least 16-18mp & 1080p video.

            • Greg Webb

              Kingyo – this may sound bizarre, but during the Football World Cup earlier this summer, I spotted a touchline photographer shooting a Hasselblad H3! Still no idea why….

            • I haven’t shot the latest gen of backs, but I have to say I’ve been consistently underwhelmed by MF backs for fashion. Even P1 C Pro doesn’t completely eliminate moire. I’ve done post on some D3x shots for other photogs and I really feel they’re stealing a lot of the fire of MF. Especially with the AF capabilities. Hassy has pulled out some interesting tricks to make their af work a little better, but it’s still crap compared to the functionality of the 51 point af or contrast detect in live view. The MF systems also suck for volume stuff because of this.

              And no, they don’t have any more DR at base ISO.

            • Jabs

              The problem with ‘readers of specifications online’ and not being aware of Technology is demonstrated by your post and others here.
              Technology moves upwards, sideways and often newer Technology eclipses the more expensive older stuff.
              E.G. – new D3100 outdoes D3S in VIDEO.
              It has a few newly developed ideas not done before and capabilities that perhaps will be introduced and PERFECTED in a D3S replacement, so think of the D3100 as a test camera or the ‘guinea pig’ in an experiment being sold to us and we the consumers doing the consumer use research FOR Nikon when we buy the D3100.
              The E.V.I.L camera is a mere fad in its’ present form, but come 3D photography, it might be great.
              Technology changes but excellence is a goal of Nikon plus others and this thrust is lost on many who use Internet BUZZ words or catchphrases.
              The D3100 is a NEW ERA of photography like the D90 WAS.
              Medium format has much better MEGAPIXELS than 35mm DSLR’s but I believe that only one Medium format digital back or camera has beaten the 35mm Nikon D3X DSLR in dynamic range and it’s from Phase.
              Megapixels is independent of dynamic range.
              Pentax in medium format is NOT there – try Mamiya who introduced a Medium format camera with 16bit RAW OUTPUT file capability and then perhaps because of this above 16bit pipeline, it will be able to do what others cannot do yet in dynamic range.
              The D3X = 16 and 14 bit, while the Mamiya = 16 and 16 bit, but the 16 bit pipeline slows down the cameras, as the current camera processors are not up to processing at SPEED and BIT structure all at the same time.
              Perhaps when they go to dual or quad processors within each processor unit (for example Expeed) and then go to 64 or 128 bit in-camera processors, then you will have also more speed and deeper bit structure like 24 bit RAW output.
              D3X = custom camera processors with deep bit structure and thus expensive.
              Medium format = huge megapixels and larger sensors = BETTER images by far.
              EVIL = ultimate convenience and lower prices for equivalent quality to the general public and NOT pros shooting either 35mm or Medium format.
              You got your terms or concepts mixed up, perhaps.
              In the future we probably won’t have camera sensors like we have now that convert light to electricity. MAYBE light to light (something to tickle or tease your mind).
              Remember the way copper wires delivered cable and the Internet and now there is fiber optic or light transmission in many markets in America? Apply that perhaps but none but God knows the future for sure, so just speculating.

            • PHB

              Well there you are picking one manufacturer and one model and finding only a few advantages. At the moment there are only two medium format backs that come anywhere close to delivering an advantage I would consider significant over the D3s/D3x. Thats not exactly impressive for cameras costing $35,000 and up.

              Most of them use sensor technology that is several years behind the curve. Take a look at some of the ‘blad ISO settings – max ISO 800!

              Sure they are great cameras, but for the money and the size they would have to be very much better than the DSLRs. And I don’t see any way that they can keep ahead as Nikon and Canon increase the resolution on their top models.

              I am expecting a 48MP DSLR with a max ISO setting of 1600 to launch for about $12,000 in the not too distant future. Whether Nikon or Canon gets there first is going to be the interesting part. You are going to have to use a big, fast lens on it to get that resolution, and you are only going to get full resolution in the center of the frame, but the MF backs have the exact same problem.

              And after mirrorless is established in the consumer market, I would expect to see an F-mount (or F-mount plus) mirrorless appearing, probably by about 2015. It is going to take them a few years before they have the format to the point you would want it on your D3, but it will get there. The mirror box is a liability for lens design.

          • lander240

            We should be noted that,
            Stop producing a camera, probably DSLR, with FF sensor
            is not the same thing as stop producing the FF sensor itself.
            The bulky DSLR is not the only box that can carry a FF sensor.
            We may not know, but SONY knows who is holding the key.

            • Anthony

              What’s the point of 48-bit output when 16-bit output is already hitting up against the variability inherent in light and sensors? Sure we can keep making incremental improvements to sensors, but we’re not going to see 48 significant bits of DR anytime soon.

            • PHB

              If they extended the dynamic range by 5 bits, there would be no need to use the ISO setting at all when shooting in RAW mode.

              But there clearly isn’t 48 bits of signal in the data coming off the sensor. About 20 is the maximum.

      • Eric Pepin

        The A900 and A850 didint cause the stir Sony expected, or anybody expected, so its quite possible. Sony is not seen as a pro photography company, pro video though yes. I can see amazing video cameras coming from Sony soon using APS sensors which of course are closer to cine 35 then stills 35 to begin with.

        • Jabs

          @Eric Pepin.
          I think that Sony already does that.
          CineAlta, I think it is called.
          They make DIGITAL movies with it but it is expensive.
          Look up Sony Broadcast and Cinema gear as most movies are shown (or projected) in digital form at Theaters in America.
          Many movies are also shot DIGITALLY on ‘digital gear’ @24 fps.
          You are late to the party – LOL!

      • elliot

        “I’ll be laughing my ass off when I see the next Sony full frame. ”

        Maybe you just need a better sense of humor. 😉

      • Anonymous

        @ Roger

        You said:

        “Tell me about it. I’ll be laughing my ass off when I see the next Sony full frame. “Sony is dropping full frame sensors!”, lol. Yeah, right! Some people will publish any garbage as news, no matter how illogical it is (hear this, Thom or Lloyd?)”

        My response:

        Do you realize that what your post can be they basis of a legal action against you? You are not just expressing an opinion but you have targeted specific persons who because of the nature of this forum, can easily be identified. This type of comment serves no purpose in this forum and instead serves to ridicule.

        You sir also need to learn to read. Thom specifically mention that Sony’s reassessment of whether to continue producing full frame dSLRs will not impact on the full frame products already in the pipe-line. This includes the two new Sony FF sensors that features improve low-light performance in 24mp and 30+mp resolution. Thom specifically also posted in his blog that he actually wants Sony to continue producing FF dSLRs to compete and to challenge Canon and Nikon to produce better dSLRs at a lower price.

        • He was talking like an idiot, but there’s a lot of that going ’round here. Either shrug it off or talk to the admin. Your own whining about legal stuff is asinine as well. Doesn’t matter who said what. Nobody cares.

          • Anonymous

            You’re right, I apologise to everyone and will no longer post here.
            I know nothing about photography and just come here to annoy others, I’m sorry.

          • Anonymous

            Whoever was pretending to be me has no effective response. Tsk tsk tsk!

            @ Micah: your response reflects on your ability to reason, and I do not even have to use derogatory words to make a point. Your use of this however reflects poorly of you.

            • Anonymous

              Hey that’s not me.
              Admin someone is pretending to be me, although since I was too stupid to choose a username and email I only have myself to blame ?

            • yes, everyone that doesn’t choose a name will be “Anonymous”

            • Admin, you should require everybody to chose and alias, at least. There anonymous and anonymous here, one in particular is trolling frequently.

            • ok, now everyone has to fill out name and email in order to post comment, this should take care of all the “Anonymous” posters

  • If Nikon is going to make us wait this long for a D700 replacement, the new FX sensor better be 20MP+ . . . Imjustsayin (please no 100 replies about how more MP isn’t necesarily better, or how Nikon’s ISO/noise performance is superior to the 5d MkII, I get all that).

    • Antonio Rojilla

      If you are so smart as to get ‘all that’, then why don’t you just get a 5D?

      • Because, Antonio, I have too many Nikon lenses to switch, and I find Nikon lenses to be superior. I rather Nikon give us a new D700x, rather than me going out getting a 5DII and $20,000 worth of lenses from Canon.

        • general

          Umm Nikon has $20K worth of lenses that are better than the Canons? Having used almost every Canon and Nikon top lens I know that there’s a problem in your figure.

          Which Nikon lenses do you have, out of plain curiosity? 😀

          • If you’ve used every lens Nikon and Canon has ever made then why is that figure so difficult to believe? I have the 14-24mm f/1.8, 24-70mm f/2.8 AF-S, 24mm f/1.4 AF-S, 50mm f/1.4 AF-S, 60mm f/2.8 Macro, Tokina 10-16m Fisheye (Nikon mount), 70-200 f/2.8 VRII, 135mm F/2.0 DC, 200-400mm f/4.0 VR, and 400mm f/2.8 VR, TC-17II, SB900 (including the SU-800 command unit), battery grips for the D700 and D300. If you were to buy all that new today it’s closer to $25,000 MSPR, but lenses have gone up in price over the years, and I didn’t always pay MSRP. So $20,000 is about right by my count. If you have used almost every Canon and Nikon top lens, I can’t imagine you spent much less.

            • Sorry, 14-24mm f/2.8, not 1.8, obvously.

    • Nathan

      I know for a fact that more megapixels equal a nicer picture. I was at the space shuttle launch in April. I shot with an 800mm on a D700 at 6.5 miles from the shuttle. I thought my pictures looked pretty darn good even when I zoomed in. Then some guy sat next to me on the bus and he had a 5D Mark 2. He used a 450mm lens at the same distance (300mm with 1.5x converter). On his camera display I was able to see stars on the flag that was on the wing of the shuttle and the words “Discovery” on the other wing. On the D700 pictures I shot I knew it was a flag and some letters, but I couldn’t discern the words or the flag. I chocked that up to more MP.

      • WoutK89

        Or resolving power of the lenses

      • Eric Pepin

        As long as you werent using a 2x converter on a 400mm which I believe reduces quality a lot more then a smaller converter but would have to try myself to know for sure. how about tripods ? did you have pro grade with proper head and plates. Was he shooting more stopped down. Id say 800mm and 12 mp should be better then cropping a 450 mm lens using 21mp unless your support was lacking.

        • Nathan

          I was using a 400mm f/5.6 ED-IF Nikkor and a TC-301. I tested the resolving power and it’s indeed really, REALLY sharp. Craters on the moon, check. Dimples on oranges more that are on a tree about a block away, check. You can’t say MP makes no difference because it has to make some type of difference or Nikon wouldn’t have brought out a 24MP D3x.

          A picture of my setup for the shot is also on that site.

        • Nathan

          Sorry, my mistake. I recall the guy saying he did have a 2x with a 1.5x with his 300mm. I remember saying “wow” when he told me what he threw onto his camera, so I guess that 900mm would contribute to seeing more.

      • PhillipG

        I would probably look to the quality of your respective tripods and lenses, as well as things like the use of mirror lock-up and a remote shutter release, before assuming the difference was due to the number of pixels in the sensors.

        • elliot

          I wouldn’t. As Thom Hogan himself said 2 years ago:

          “… could you do better than the 12mp sensor does today? Almost certainly. 12 to 16 is marginal in resolution, though. I generally say you need a 20% change in pixel count across a dimension to have clear visual differences and 16mp wouldn’t get you there.”

      • Just A Thought

        To get double the resolution of your 12MP camera you need four times as many pixels. so you would need a 48MP sensor. I suspect other factors contributed to the difference you saw.

        You shot with an 800mm lens, which means it was possibly a mirror lens. If it was a rare super Nikkor Telephoto then you would likely not be riding on a bus (limo maybe). The Canon user shot with a 300mm prime and likely a matched 1.4 teleconvertor. Mirror lenses can be very good but they are no match for an all glass lens. Contrast affects perceived resolution and photos from Mirror Lenses lack contrast as compared to the same shot taken with a good quality all glass lens.

        Now, if you shot with a 400mm and a 2X tele-convertor then that explains why your pics lacked detail. It has nothing to do with pixel count.

        At 800mm camera shake is also a very big problem – even mirror flap could cause problems. Not sure if either of you took photos with the mirror raised and locked and with a remote release. Probably shot a highest frame rate available on camera = more shake issues from bouncing shutter

        You might have been better served to have rented a Nikon d300S and a 500mm Nikkor super Tele lens. The crop sensor would give you 750mm and high end Nikkors are hard to beat. Just don’t forget to also rent the super heavy tripod, gimball mount and remote shutter release.

        Or you could save yourself a lot of money and hassle and buy some amazing photos from Nasa. There was a Nasa photog who used to post on DPReview – forget his name, but his shots were like eye candy. You could not come close, no matter how expensive your gear was because he had access.

        • Nathan

          Check out my link and see the picture for yourself and the one of the moon while you’re at it. It was a prime lens. There’s a picture of my camera setup on a 5 pound Gitzo tripod. I don’t think stability was a problem.

          I made a mistake and the guy with the Canon trumped me with focal length by using 2 teleconverters on his 300mm lens. One was a, as you say, 1.4x (or 1.5x (I don’t know Canon stuff)) and a 2x converter, so I guess he had an 840mm lens. His picture with 21MP still had more resolution than my 12MP could muster. So we can talk about this till we turn blue in the face, but I saw his picture and I was able to read “Discovery” and see the flag in full glory. Look at my 100% magnification shot. It was shot at 1/1600 so the camera wasn’t shaking.

          • WoutK89

            so it must have been out of focus, as I can tell by the foreground in the picture.

      • Jabs

        Yes more megapixels do count, but perhaps your problem was a phenomenon called ‘atmospheric haze’ that comes into play when you go over a certain mm size (focal length) in lens. You also used teleconverters and these make the problem worse by magnifying the very problem.
        Generally above 400-600 mm and above in focal length, you have problems like that. The other shooter used a shorter focal length which did not cause that problem. Next time try a circular polarizer for glare or atmospheric haze reduction, sand bags on your tripod if needed and maybe a higher shutter speed or lower/higher f-stop and ISO combination to give you the exposure you need. ED glass minimizes atmospheric haze but glare or veiling flare is reduced by better lenses of the focal length that you need plus use the Nikon drop in polarizer if you had the 400 F3.5 and if you used the 400 F5.6, then you need a front circular polarizer, if I remember right. You rotate the circular polarizer to reduce veiling flare or glare from the flames/ground swell vapor and then compensate in your exposure (usually + in 1/3rds).
        Sounds like atmospheric haze from you being so far away was photographed due to the longer focal length that you used – a typical problem, hence nano-coated newer ED long lenses by Nikon. I used to deal with that issue with TC’s and 50-300 F4.5 ED plus 300 F4.0 AF ED at times when shooting from long distances and nothing seems sharp or even in focus. It is similar to shooting through a fence with a short -vs- long focal length lens. Depth of field issues made worse by the longer focal length.
        Have a good one!

        • For super far away objects I’ve used my D300 instead of D700. The 1.5 crop factor can help and to the extent any “atmospheric haze” issues may exist; I assume they are not affected by the crop factor of your camera? Anyway I’ve frequently shot with my D300, 400mm f/2.8 VRII, with a TC-17II with no problems. The 400mm f.2,8 VRII is in my humble opinion the sharpest telephoto lens offered by Nikon. As for whether 20MP makes a difference; it does to me since I sell stock photos and anyone selling stock photos know the sizeof your images can make a difference on the popularity/margin of certain images (not all, since not all stock images are meant to be blown up into large prints).

          • Jabs

            The key factor is what lens they both started with as the base lens.
            One started with a 300 mm Canon, if I remember and the other started with a 400mm F5.6 Nikon.
            The Nikon 400 mm F5.6 is an old design and thus when you add the TC-301 teleconverter, you have issues. The TC-301 is a 2X converter, so now they have a 800 F11.0 minimum aperture as in SLOOOOW!
            Look here:

            That is one source of the problem and it is a known fact that older Nikon or even other brands were all subject to atmospheric haze especially when shooting the Shuttle in Florida as there is ocean or sea there. The lens that you start with is now degraded by this OLD TC-301 back from the days of AIS manual Nikon lenses and the 400 F5.6 is a real old design which came along the time of the better 400 F3.5 ED which was superseded by the various 400 2.8 ED-IF and the later variants.
            Hence old glass and old teleconverter being exposed by a modern camera. When you add any teleconverter to a lens, the results are less stellar than the lens by itself.
            You probably are spoiled by newer lens – lol.

            • I see. Yeah when I can avoid the teleconverter I do. So instead of shooting a D700 with the 400 f/2.8 VRII, if I need more reach I just use the D300. The AF speed is also can be significantly affected by the TC. the 400 f/2.8 VRII is one hell of a sharp lens I have to say.

            • Jabs

              Yeah – the formula is – when shooting over a distance in desert heat, over water and over long open distances – you get an effect that is caused by rising heat waves, water vapor rising naturally from the Earth and such items which the naked eyes often do not see or even focus on.
              When you look at the results on film or on digital, you now see the effect often called veiling flare and even an effect caused by atmospheric haze. You note this in high humidity areas, near seas, beaches, desert areas, etc,. Heat rises – Water vapor rises, dust acts like a mirror often and sand being silica, acts also like a mirror or reflector and messes up clarity or even things that the naked eye compensates for or overlooks, now becoming frustrating clear on film or digital. It is like shooting stainless steel or chrome. Hence, my recommendation for polarizers and even UV filters. I used to use a combination of Nikon A2 filter and a Nikon or B+W Circular polarizer and then rotate the polarizer slightly for the minimum effect needed to remove glare and bring clarity. A2 reduces blue tinges – (chromatic opposite of blue) and the polarizer reduces the mirror like reflection of the long distance shot that the long focal length also magnified. B+W also had a few combination warming/polarizing filter but they were expensive. I had one for a 50-300 F4.5 ED and at 95mm filter size, it was huge and expensive. Then I would use a 95 L37C Nikon UV filter and the B+W combination stacked, plus a TC a few times (for reach) with Fuji Velvia 50D, 64T or 100D Pro (or even 400D) emulsions on an F3HP/MD-4/MN-2 combo mounted on a Gitzo tripod and my camera bag hung below the lower tripod column to act as ballast – all to get long distance shots.
              Over the years almost everyone complains about shooting the Space Shuttle from Cape Canaveral and the further away you are, the worse it becomes as you said maybe you were 6+ miles away, right.
              The Canon shooter’s base lens was 300 mm and yours was 400 mm, hence you started out with a disadvantage – meaning – your focal length choice magnified atmospheric haze greater than his lens.
              Using the DX body now magnifies things within the camera (better) and not within the external lens/TC combo, so better results as you took the TC301 out of the picture to gain focal length. Yeah – 400 F2.8 is really great but you also started with F2.8 there and not the slower older F5.6 of the older 400 F5.6 and you probably approached its diffraction limit as when you added the TC 301, your maximum F-stop was now F11 – ouch! You probably did not factor in the doubling of F stop as I don’t know if the D700 compensates with the older teleconverters such as the TC 301. Look at Nikon’s web site or look at your camera – LOL! If you shot @ F11 with the TC 301 attached, you were actually shooting at F22 and also if you chose F16 = F32 now. Some Nikon bodies compensate by EXPOSURE with TC’s but do not tell you at what F-stop you were shooting.
              Was the 400 F5.6 AI or AIS – the ED version or not?

            • Nathan

              Actually, yes, I did take into account the 2x f/stop. I had a f/5.6 to start and used the lens wide open so I could have f/11. I used ISO 3200 because I wanted 1/1600 so I wouldn’t get any blurring. I went back and forth with this over and over again when preparing because i knew I could get away with using 1/1000 and a lower ISO, but I wanted to be sure there was no blurring. I was successful.

              As for the person that mention the picture is out of focus, it isn’t. Since you don’t have the raw file you can’t see it, but at 200% and 400% viewing of the picture it’s as sharp as a tack. I’m able to see the individual windows and the lines that separate the booster tanks. Also, if you look closely you can see the phone lines coming from the telephone poles. If the photo was out of focus you wouldn’t see that. I had two hours to sit there and focus and refocus and refocus again to make sure that the shuttle was in focus.

              As for atmospheric haze, there was some, but not much since the air was basically still, the temperture was around 65 degrees and humidity wasn’t too bad. It was a picture perfect morning.

              The photo was as good as could possibly be made with the equipment I had. I’m usually the first to point to any flaws in the system and the only flaw I found was that I couldn’t stop the lens down to f/16 (f/8) without losing the shutter speed. I wasn’t about to shoot without the TC-301, which does a fantastic job since it was matched for the telephoto lens.

              Truthfully I’m really, really surprised of all the people that trash the older lenses and say they are terrible. I’m assuming that many of you that do poo-poo the lenses haven’t ever tried them, thus thinking your 80-400 zoom lens is superior (also a f/5.6 lens by the way). Also for those that think that everyone is rich and can afford the latest 200-400 f/4 or 400 f/2.8 lens, guess what, we’re not all rich. My lens with teleconverter cost less than $1000 and I feel got the picture I wanted. If you think differently, then great. I got what I wanted and to me that’s all that matters.

              As for the Canon making a better picture, more megapixels do give a more detailed picture, especially when pixel peeping.

    • lolcatmasterFTW!!!

      Yes mostly because most of those asking here for a 20mp full frame sensor will be printing HUGE stuff!!!!… oh right… they do not even print 8×10 inches… much less larger…

  • Ant

    I guess the big question about this report is whether Nikon have a Fab with the capacity to efficiently produce consumer-volume DX sensors. I can understand Nikon designing and developing the sensor, but it would make more sense for them to get a 3rd party (Sony for example) to manufacture the thing. That might explain some of the Nikon / Sony confusion about the origin of the D3100 sensor. I don’t know enough about Nikon’s CMOS capacity to make a sensible judgment, maybe someone else can take the reins on this.

    • There’s plenty of extra or underutilized fab capacity that can be bought these days. The real question is yield. For an entry level camera that will sell probably a million units in its short life you need a predictable and high yield, otherwise you risk announcing a camera that you can’t make in sufficient quantity to meet demand. Obviously, wherever this sensor is coming from Nikon seems confident that they can produce it volume.

  • peter

    Is Nikon being forced down this road as an effort by the giant Japanese parent companies in an effort to drive them to be taken over by another? In Japan companies are still run like warring factions and they do it on a VERY large scale. I would hate to see Nikon swallowed up because they are pushed out of their core business and into bleeding edge electronics R & D.

    • Ant

      I don’t believe so. Whilst Sony and some other massive Japanese electronics companies are indeed massive, their different divisions tend to be run as silos. The Sony Imaging operation is managed separately to the Sony Semiconductor operation for example. From the point of view of Sony Semiconductor both Nikon and Sony Imaging are customers. Nikon have the bigger market share and volume, so if anything Nikon are the more important customer for Sony Semiconductor.

      • I agree with this. The big loser here would seem to be Sony Semiconductor, who may have just lost their biggest client in the large format sensors.

  • Makoto Kimura

    the 17mm lives on..! come on nikon you can do it, be a pal & all that. i luff you

  • I love conspiracies 😀

  • steve

    Also don’t forget that nikon also makes equipment for fab’s

    • elliot

      Kind of irrelevant. Nikon don’t have the needed billion-dollar fab factory (or the way, as Sony do, to cost-out production to multiple buyers) to mass-produce their sensors efficiently and cost-effectively. Of COURSE they job it out to someone else, and in this economy there are multiple partners, including Sony, Dalsa and Kodak, with space on their existing lines for custom chip production.

  • Dan Spy

    If Sony wont make another FX sensor, there are at least 6 other electronic manufacturers than can. Probably many more.
    Thats not a problem for Nikon.Not at all.

    • If I had a full frame sensor design, there are plenty of places I could take it to have it produced. The issue here isn’t about whether you can make a full frame sensor, but whether there is any economic incentive to do so. Canon and Nikon are the only two companies currently making enough full frame bodies to justify the extra expenses of running full frame sensors on a fab. Without Nikon, Sony probably doesn’t have enough volume to run a machine dedicated to full frame. In other words, it isn’t Sony denying Nikon full frame sensors, it’s Nikon now building their own full frame sensors that’s impacting Sony Semiconductor.

      While the Japanese companies think longer term than American or Western firms, they too think about ROI (return on investment). For Sony, the critical discussion is whether the ROI on full frame is worth pursuing. Put another way, if a company has X dollars to invest, where to they get the most return? For Sony Imaging, the only thing right now with interchangeable lenses that is showing any signs of breaking out is the NEX.

      My impression is that Sony has fumbled a clear advantage that they had early on, when they were one of the few sources of quality sensors. They never built and supported that advantage to dominate small (e.g. cell phone) sensors, and now they are being marginalized at the top, too. Demand for compact camera sensors, which Sony dominated, is flat and will decline. Demand for APS sensors is slightly up, but Sony appears to have lost their biggest customer if Nikon is indeed now doing their own DX sensors. Demand for full frame sensors also appears to be lost with Nikon going their own way, there, too.

      People have to separate Sony Semiconductor from Sony Imaging, as they are separate companies within the Sony umbrella. Sony Semiconductor appears to be losing ground and customers. Sony Imaging is not a large enough customer to make up for that, so Semiconductor has some tough choices ahead, it appears.

      • Carl

        Couldn’t Sony Imaging move FF sensor production from Sony Semiconductor to a Nikon fab? Devoloped by Sony – produced by Nikon, just the other way around than before.

        • T140Rider

          do you have any idea how much it costs to setup a Fab these Days?
          Billions of $$$.
          The Dresden Fab nearly bankrupted AMD.

          Sensors are another kettle of fish altogether. Far more specialised than X86 CPU’s.
          I doubt that Nikon had the $$$, Yen or whatever to sink into setting up their own Sensor FAB. Besides, they take literally years to build.
          First you put up a building and make it clean. Then inside that you build another cleaner building. We would have know about these plans for sometime.
          If Sony are no longer making the sensors for Nikon then they have simply moved their FAB to someone else. Perhaps Fujutsu? In case you don’t know it, they make the majority of Sun/Oracle SPARC Cpu’s these days.
          There is currently a huge amount of unused FAB capacity around the world but not all of it can produce Camera Sensors. A lot of it is for 65nm and above technology. Intel & AMD are looking at 32/25nm for their next gen CPU’s but even Intel is having to think hard about the cost.

  • Adam

    More MP bigger the file.. I rather have an FX at 14 -18 mp. Also, since there is a new sensor, I don’t think I’m going to be the first one to buy it, I will take the cautiously-optimistic route and wait.

  • peter

    In a way I guess it is a bit paranoid. I work at a small community college that has been run by an out sourced company for a couple of years now and we are all looking for the monster under the bed. If indeed these divisions are run as silos it would seem like much ado about nothing. In America we’ve been seeing a lot of take overs with the rich getting oh so obscenely rich and it is usually at the expense of tradition and quality. I very much admire the Japanese for holding onto “national treasures” and as Nikon rose from the ashes and still maintains a profoundly good product, I guess I was wincing at the possibility that another transcendent company could see peril on the horizon. Yes it would scare the hell out of me if my sensor supplier said, “we’re done”.

    • Eric Pepin

      when your sensor supplier is an even bigger more popular japanese company your probably good though. If nikon can order in enough volume im sure Sony would have no problem making a healthy profit on manufacturing sensors for Nikon. Dont sweat it.

      If a demand excists, a product will fill it.

  • Roger

    Come on people, why is this news?!!!


    • Roger, I have been monitoring Nikon news daily for almost three years and I have not seen any other public announcements from Nikon on that topic except the two I mentioned above. Please provide us with a legitimate link.

      btw, using CAPS on the Internet means that you are screaming

      • El Aura

        is a link on legit enough?
        Nous souhaitons utiliser nos propres capteurs dans les reflex plus populaires [petits capteurs APS-C, NDLR], car les performances de nos capteurs sont meilleures. Néanmoins, cela prendra un peu de temps car il faut faire des économies d’échelle.
        In short, they want to use their own sensors in their high-volume DSLRs (ie, DX) but it will take a little bit of time.
        How is that not announcing that they are moving towards dropping Sony?

        • ok, you got me – I did forget about that one, even though I covered it here. I will update the post. Respect 🙂

          • Matt XVI


            I just wanted to tell you how much I love this site. It makes me so unproductive at work as I literally check it about every half hour and get caught up in the comments thereafter.

            Just wanted to say thanks and keep up the good work!



      • Achille

        Dear Admin, I agree with Roger
        Remember, when the D90 just went out, there were some discussions around, about the sensor: It appeared that the D90’s sensor was not from sony, like the D300 one, but from Renesas. That’s why some guys ( not sure if Nikon said it itself) argued that this was a proof that Nikon was seperating from Sony to build their sensors
        I wonder why everybody forgot this story, It was not in my dreams right?

        • PhillipG

          According to DPReview and Thom Hogan, the D90 sensor from Sony.

        • BornOptimist

          Parts of this was probably in your dreams, because it was speculation around the D3 sensor that it was/is manufactured by Renesas. There has also been some discussion about the D300 sensor (14 bits while the Sony is 12 bits).

    • elliot

      Prove it.

  • Jason nikon

    So nikon new 16 mpx sensor is made by them do you think it will perform better then a Sony 16 Mpx eg prosesing and power comsumsion

  • Wierdo

    I think this is why the rumor and the leak are not all over the place, Nikon have tighten their security by making their own sensor…
    Im sure it will perform as great or better than sony sensor, they must’ve develop and try making it even before the D3x was out, if not how will they back up their quality standard…
    Im sure their not stupid enough to make something thats awful, although they are stupid enough not to meet supply demand or DSLR replacement for all the people that been crying about it… ;P

    • I think you’re right. Many of the Nikon leaks come from suppliers, not Nikon themselves. But remember, if Nikon is using an outside fab, they still have a supplier, perhaps maybe not as leaky a one as Sony Semiconductor was ;~).

  • Revenue Matters

    i could be wrong but everyone clamoring for nikon this that or the other miss one very important factor in what nikon does or does not. that is revenue numbers? money is made not by $6000 cameras but those people between $300 and $1300. those people who love taking photos but never will sell a photo. (too many peter parkers out there already). so who cares if nikon makes their own sensors, who cares who is crying over a D4, D700 etc…as long as revenue shows that nikon is doing a stellar job then that’s what matters

    • PhillipG

      I’m also happy if Nikon is making money on their entry-level gear, because it means more money to put into R&D of their higher-end cameras and lenses.

      If Nikon will be using only sensors they design themselves, it’s interesting from a gearhead & customer point of view because it means that

      a) it may be harder to predict what sensors their new products will use.
      b) there may be faster iterations of improved sensors within Nikon’s product line

      Look at the difference in the upgrades from D300->D300s and D3->D3s. The D300 and D300s use a Sony derived sensor, and the upgrade didn’t enhance the sensor. The D3 and D3s use a Nikon-designed sensor that was significantly upgraded for the D3s to have less noise and greater dynamic range at higher ISOs. As a Nikon customer, which sort of upgrade do you prefer: D300->D300s or D3->D3s?

    • Missed the Spider Man reference for a second and thought that “Peter Parker” was something… vaguely illicit. 🙂 Must be bedtime.

  • The invisible man

    That make sense, if Nikon release the 3 sensors DSLR, it may be oblige to produce them itself.
    I think 2011 will be a BIG year for Nikon.

  • zzddrr

    Guys, I am in trouble!!! Since Nikon passed magical limit of 12Mp with this 14Mp D3100 and if it is in fact true that Nikon produces its sensors pretty soon I cannot even bitch/complain about anything. 🙂 I knew that one day Nikon will screw me up 🙂

    • venancio

      oh, there’s still the video side, or F-mount versus K-mount, or even night vision, but then again, they must have thought of these ages ago… maybe anonymous has something for you…

    • Enesunkie

      Ah now don’t worry, it’s Nikon, there will be lots of opportunities to complain about lacking or gimmicky features!

    • zzddrr with nothing to complain about? I do believe it’s gotten cold in here!


      • zzddrr

        Fried Toast, I am shocked. The day is getting closer when I no longer can cmplain 🙂

        • Ok it’s true. Hell froze over! jk zzddrr

  • Zeiss-Ikon

    And so it begins.


  • Anon

    Didn’t the Nikon Demark guy in the interview state he thought the sensor in the D3100 was a Sony sensor?

    • Yes, he did. I’ve heard different things from different parts of Nikon. As usual, the only sources that seem to know any details all seem to be in Japan. The subsidiaries only get information about the technology from corporate in Japan. What seems odd to me is that, if Nikon really is making a sensor split from Sony and doing their own DX sensors now, that Nikon didn’t make a bigger splash about that. Perhaps they’re saving that for a day in the future when they will suddenly say “all Nikon cameras use Nikon sensors.” I don’t know. But historically, Nikon has been both poor at communicating and marketing their technological advantages and changes. And they’re incredibly poor at building the basic marketing case of technology->user benefit.

      Take Nano coating, for example. Does the average user know what the heck it is or what it actually does? Nope. To most people it’s just some geeky term that’s suddenly appeared on product specification lists ;~).

      • Sahaja

        Sounds a bit like Toyota who got into a costly mess partly because they were trying to manage everything centrally from Japan.

      • maybe they will make the “big splash” after the D90 replacement is announced in 3 weeks

      • The Man from Mandrem

        The average user can’t afford the damn thing. I would think at this point people just trust a Nikon with a new descriptor and a higher price tag is better or wait for the lense tests…

      • Roger

        Maybe they’re not parading around the “sensor is made by Nikon” because they did so with the bad old D2h, and we all know what happened there. Hell, they even had to change the naming, they had to ditch the “h” just so that D3 wouldnt have anything to do with the D2h.

        And hell yes, they’re poor at marketing. Forget nano-coating, it gets even worse – average user out there thinks 5D Mark II is just as good as the D3x, just cheaper. 😀

  • Phil Koenig

    There’s a big difference between “developed by company X”, versus “produced by company X” or especially “manufactured by company X”.

    Many “chip companies” today are “fabless”, which basically means that they do the design and marketing, but not the manufacturing of their product.

    Given that the capital investment and expertise required to produce silicon chips of various sorts (including imaging sensors) is non-trivial, and given that that is far from Nikon’s core competencies and core businesses, and given that Nikon is a much smaller company than Panasonic, Canon or Sony without their electronics-manufacturing legacy, I don’t think it’s very surprising at all that they don’t actually manufacture their own image sensors. (Caveat: Nikon is part of the Mitsubishi group, which may have some resources of their own to bear in that area, which I’m sure Nikon draws-upon from time to time.)

    I doubt that Pentax (Hoya), Olympus et al do either. Panasonic and Sony are a special case because they have substantial chip-making capabilities, and Canon has a much deeper background in electronics than Nikon does.

    Nikon started out as an optical manufacturer, and that remains a key differentiator between Nikon and most of their Japanese competition. (Canon’s first cameras actually had Nikkor lenses on them, because Canon didn’t have that capability when they started out, whereas Nikon had been building optics for decades before Canon was founded.)

    Other prestigious optical companies such as Leica don’t make their own sensors either.

    I would however find it perfectly plausible that Nikon has reached the point where they are heavily involved in the design process of the sensors, and are having their sensor OEM (quite possibly Sony) building them customized versions of the chips that Sony uses themselves and sells to others. Just because a particular sensor has size “X” and has “Y” number of megapixels on it, doesn’t mean that it’s 100% identical to the chip used in someone else’s device. Also, Nikon clearly does do their own image-processing tech, which also has a substantial impact on the ultimate image quality their cameras produce.

    Certainly Nikon understands that being beholden to a competitor for sensors is a potentially dangerous situation to be in these days, so they will undoubtedly do what they can to differentiate themselves from the market, as noted above.

    On the other hand, companies like Sony, Samsung et al also need to sell lots of sensors to make that business viable (one of the reasons I doubt Nikon wants to get wrapped up in that business), and unless you have a 60% marketshare in the products those sensors are getting used in, you will need to find customers for your chips. If your potential customers get the sense that you are trying to use that sensor OEM relationship as a trojan horse to undermine them in the marketplace, you will most likely quickly discover that you no longer have any customers.

    • Phil Koenig

      I forgot to mention in the previous comment that another option for Nikon is to have someone like Sony or Panasonic (former Matsushita), Toshiba, Samsung etc manufacture chips to Nikon’s specifications.

      To some extent the OEM still has to have certain core capabilities, ie if Nikon wishes to do CMOS chips and the OEM has no expertise in anything other than CCD, then you need to find an OEM who can do what you need them to do.

      Bottom line is, there are lots of grey-areas between “buys standard off-the-shelf parts”, and “specifies, designs, builds and owns all manufacturing facilities for all components. (otherwise known as vertically-integrated)

      And yes, Nikon makes steppers which are used in chip production. That has nothing to do with whether they have the resources, capabilities, and interest in becoming a chip manufacturer themselves. (I don’t think any stepper manufacturers are also in the chip business.)

    • Just a point of clarification: Nikon has been formally involved in sensor research and development since at least 1988. Early on they were present at many industry conferences and even occasionally published papers or patents. Nikon also hired a number of key CMOS researchers early on, so they were certainly exploring CMOS sensors very, very early. Sony originally didn’t want to do the APS sensor, but it was Nikon that pushed them there (and that original D1 sensor was actually a 10.4mp sensor that was binned to produce better results, something that I don’t think Sony had done before).

      • ELK

        Well done, Nikon! Of course using Sony sensor was helping to reduce costs, but now Nikon will have more flexibilty. For instance they theoretically could use Sony sensors for more video oriented bodies, and Nikon sensors – for photography oriented bodies. Also Nikon production cycles won’t heavily depend on Sony’s. I’m a Canon user and seeing that Nikon makes right moves recent few years makes me very glad, since Canon also won’t have a chance anymore to be too lazy. Of course if Canon doesn’t plan to forget DSLR photography for sake of DSLR videography 🙂

    • Jabs

      @Phil Koenig.
      Great points!
      A summation from me, perhaps.
      Sony caused quite a STIR in the Photographic world, when they became both Sensor Fab Producer and 35mm Camera Producer, when they bought Minolta. They now are direct competitors in finished products to those whom they also supply components to. In other words, they get to see their competitors designs when they produce them in their Fab.
      Hence, you have to leave them or they have to restructure like AMD and Global Foundry did. Maybe they (Nikon and Sony or Sony only – to bar conflict of interest and competition with itself or clients) already did this.
      A very interesting discussion here.

  • SimonC

    About time they start designing their DX sensors. It’s probably a coincidence that they have spec’d the same MP as Sony. Still, it is a mystery where they have the sensor produced.

    Let’s hope the D3100 has some of the “secret sauce” Nikon put in the D3s for high-ISO performance.

    • No, it’s not a coincidence at all that they’d spec similar to Sony. Consider what would have happened if Nikon had failed to get their sensor out in a timely fashion or that it didn’t deliver expected results: the camera using it would never appear if they didn’t have a backup plan. I’m betting that the backup plan was always “we can stick a Sony sensor in there if we have to.”

  • Tim

    Didn’t Nikon claim that they developed their own sensor as far back as the D2h?

    • SimonC

      Ah, yes, you’re right – they did develop the LBCAST sensor for the D2h.

      It’s good news that they have started to put their own designs into entry level models. This means that the D400 will most certainly take on D4 tech. (but at smaller size).

    • Roger

      Yes, but luckily for us – their recent sensors have been MUCH better than the infamous D2h sensor.

      It’s what happens when companies listen to users. Nikon users said they didnt need more than 4mp, so Nikon made the 4mp D2h and then what happened? All those users simply went and bought the 8mp 1D Mark II instead.

      So anytime Nikon officials hear their users saying “we dont need more than 12mp” or “D3s noise is good enough, we’ll never need anything better” or my personal favorite insanity: “We need VR in every lens, including 16mm Fisheye”, what they should do is cover their ears, and start running away from those completely insane and deluded users.

  • Chuck

    LOL, Nikon don’t have a dam fab, these cost big bucks and you need huge volumes to fill them. I guess Nikon secretly built a fab… yeah right.

    But maybe if they did it could expain the shortage of the D3S sensor as Nikon might be having problems ramping that secret fab of theirs.

  • brave new world

    I don’t understand the discussion: Nikon can design their sensors and other chips. Production can be done in several fabs. Nikon is one of the most important suppliers to fabs (their line of business is very successful). conclusion: Nikon has potentially several dozens of working relationships with fabs. Consequently: they could approach any of them on a level that is different to those who “just” have a design but no other business-relationship to the fab.

    regarding 40+ MPXL sensors – once you’ve used such equipment, you will immediately see the difference. It is a different market and the best is to rent equipment to find out …

  • le_eiji

    Notice why they say “developed by Nikon”, instead of “produced by Nikon”. They use the same rhetoric when they released D3x which was entirely designed by Sony, peppered with a Nikon flavor. That’s the answer to your concern. It is made by Sony, developed by Nikon. There is no contradiction in saying that they are “developed by Nikon”. As a matter of fact, Sony as well as Canon has far more advanced semi-conductor technology than Nikon.

  • There are two different new rumors about upcoming FX. Which one would you purchase?

    Camera A: 24 mp native iso 200-6400 exp. to 25K

    Camera B: 16 mp native iso 200-12800 exp. to 100K

    Both will be in the form of D700 with everything being equal incl. frame rate. Both will accomodate 1080p video. Prices will be equal too.

    Image cleaniness is about the same with both cameras up to about iso 1600. Camera A will boast 1 stop better dynamic range, while Camera B will be 1 stop better at very high sensitivity settings.

    I’d prefer Camera A without any doubt, because resolution & DR is more important for me. Iso above 1600 is a rare need. But I know there are different applications.

    I believe the preference between these models will be about 50-50% if not 60-40%.

    • Personally, I’d go with B because I currently shoot in the dark far more than I shoot landscape. However, that could change if I move and/or change jobs. So it’s not an easy answer as I could easily move to the A group within a year’s time. Currently, have no need for 12mp, but don’t begrudge those that want more (just those that whine about it incessantly).

      Re: the MF comments from above, I’d be interested in the difference between MF and the D3x (as well as other lower-on-the-chain Nikons/Canons). I played with a Phase One digital back not long ago, but it was in a dark room, so it was more of just getting the feel for it, rather than seeing what sort of performance it had.
      Guess I could rent one, but not sure I’d know how to turn the thing on 😉

    • jimmy

      Camera B would – in theory – have a much higher fps, which is very important to me for sports.

    • You forgot that if calibrated ISO range has 7 stops (200-12400) instead of 6 (200-6400), actual DR at base ISO is likely better for the former, meaning that detail in darker areas are possibly easier to recover. Thats why D3s beats every other camera in that respect, including D3x.

  • Abhinav

    I will not care whether nikon uses Sony sensors or nikon makes them on their own :D.I only want excellent performance 🙂

  • brave new world

    14 MPXL for DX means at same pixel-size a 28 MPXL for FX.

    When can we expect processors in the cameras having sufficient compute resources?
    It would be great to have a comparison of Expeed vs. Expeed2 processors to estimate, if larger sensors will need 2 or even 4 processors for high end.

    If I had two wishes – at 16bit no tweaked 12bits being exposed as 14bits … :
    1.) 28 MPXL at similar sensitivity, noise level and 2K video
    2.) 38 MPXL having a bit less sensitivity, same noise level and 4K video plus 2K video at 120 fps

    That would be the fair alternatives to Phase1, red and finally Hasselblad

  • brave new world

    I wouldn’t go that far … allow a reserve, if the sensor is getting bigger … 31 minus 10% gets you to 28 …

    anyway: it is significantly better than 12 or 24 MPXL

    regarding previous posts: “camera a” please including a stereoscopic system to use two at a time in full sync (stills and video) to produce 3D …

  • Nick Kahn

    Regarding potential scale-up of the 14.2MP D3100 DX sensor to FF:

    The sensor size is 23.1 x 15.4mm (which is a little smaller than other Nikon DX sensors). The D3X sensor is 35.9 x 24.0 mm. That makes the sensor crop factor about 1.554X. That’s also about a 5 micron photosite. The D3X has about a 5.9 micron pixel and the D700 is up around 8.4 micron I think. Still, let’s say that Nikon improves the noise characteristics and efficiency and can give good IQ using a 5 micron photosite. Then, scaling up to FX based on the pixel density of the D3100 gives

    14.2MP x 1.554^2 = 34.3 MP

    So we might expect a 34 MP FF sensor if Nikon chooses to scaleup the D3100 sensor. I would guess they would use larger photosites but then they may improve noise and efficiency enough to use the 5 micron size. Doubt they will go lower.

    And for the folks who think that lenses will not limit resolution when using the higher MP sensors, I respectfully disagree. Looking at photozone’s full frame resolution numbers, which are based on Norman Koren’s Imatest program, the highest res lenses I have seen are the new 70-200/2.8 and the 200/2 with each just slightly above 4000 line widths per picture height (LWPH) (PH = sensor height). Those results are based on the MTF50. If you have lower contrast in an image, you probably will not get that resolution. You need two lines (line widths) per line pair. The conversion to line widths per mm (lp/mm) is

    Lens lp/mm = LWPH/(2*PH)

    so for a LWPH of 4000 and a PH (sensor height in mm) of 24, we get about 83 lp/mm, for the very best lenses. Maybe there are some that are better, but I am quite sure that there are a lot that are worse. Human eye resolution is usually stated to be for best case around 5-6 lp/mm. Let’s call it 5 lp/mm for very best case and for people who want to stick their noses on a print (12 inch viewing distance) like you might do for the old large format master’s prints to see the fine detail. Personally, I think we can all live with 3 lp/mm (that’s 6 dots per mm) which is tough for me to see. Still, the old publishing standards gave 300 dpi and you need 2 dots per line pair, so that’s 150 lp per inch or about 6 lp/mm. That’s why people say to print at 300 dpi for max res on the print, best case.

    If you take the best len’s resolution of 80 lp/mm and you want let’s say 5 lp/mm on your viewed image or print, then you can enlarge the sensor image by 80/5 = 16X, if the sensor MP count is high enough.

    The 24MP D3X is 4000 x 6000 pixels and allows enlarging to about 16 x 24 at 240dpi, which is a 16X enlargement of the FF sensor. So, the best lenses are also allowing 16X enlargement conveniently enough. However, if you use a lesser lens, say the 60 lp/mm lens, then you can only take advantage of about 0.75 of the 24MP sensor resolution. This is not even taking into account lens or camera shake.

    I know there are plenty or people who like to say ” I can print at 30 x 40 inches from my D300 and they look great!” but they have uprezzed which is making up data and they are really not printing to show the highest resolution that their lens and camera can capture. Don’t confuse smooth tonality (in this case I mean lack of visible dots on the print) with resolution.

    So, while I too want a high MP cam one day, I recognize that I need very good lenses and technique to have any chance or realizing the max possible resolution. Larger format cameras do not enlarge as much and so do not need as much lens resolution to begin with and generally the net result, with enough MP to start with, is a higher res print, but I think they would need 60 plus MP and great technique to be able to see a difference, and then printed very large.


    • Denko

      Thanks for the writeup.

    • FakeKenRockwell

      Nick, this needs to be permanently linked and archived on NikonRumors. Thanks for the cogent post on resolution and MPs.

      • Nick Kahn

        No problem Denko. I think it is good for people to understand the limitations of their current equipment and even of potential new purchases. New and better is good. We all need to consider our lenses and technique (specifically wrt minimizing vibration and shake) though to produce the highest resolution images, if that is a goal and not all photography and perhaps maybe even most, is not about highest res.

        Thanks FakeKR. If I knew how to link and archive I might do that. Do you perchance have a fake site with this info? If so, I will be happy to increase your fake traffic to find out ;>).

        • FakeKenRockwell

          no, not fakeKR site at the moment but it’s very tempting

    • Jabs

      @Nick Kahn.
      Great post. Can I ask what field you are in?
      Just curious – you sound like an Engineer and/or an informed and logical thinker – LOL!

      • Nick Kahn

        Jabs: Thanks, and you are close on part 1, questionable on part b ;>). As to your question, Nick Kahn must exist now only in our imaginations.

        Unless FakeKR recruits me to write on his fake site.


  • I was first here commenting sizes didn’t match (with Sony’s 14MP sensor in NEX3). I posted same remark at DPR and this poster checked images of sensors, comparing D3’s sensor with it, remarking connectors looked different. I posted it here 2 days ago, here’s the link to his DPR post.

    I think this is big news for Nikon users and for the industry in general. But likley it’s made in some plant outside Nikon, I don’t think they have their own fab unit, so they?

    One relevant point is that base ISO is now 100, and highest calibrated ISO is 3200, thus there are 6 calibrated ISo stops. This is the first such DX sensor, only D3 and D700 hve that. D3s’ shows 7 calibrated stops (200-25600), even though some may question if the 25K is actually calibrated.

    If D90+ is 100-6400, calibrated, that’s D3s’ tech level, even though my guess is that it’ll gain an ISO at top because of sensitivity, with base ISO at 200.

    • Nick Kahn

      Hi rlhp: It will be good to see the reviews. Could be impressive IQ for such a low priced camera and make it a great bargain.


    • Correction: D3s’ range is 200-12800, 7 stops. There are 3 uncalibrated levels. All other comments and numbers are correct.

  • Tor Gunnar Berland

    One of Nikon’s biggest customer is Intel; so I think Nikon is capable of producing their own sensors if they wanted.

    • rhlpetrus

      one thing is to produce steppers, which they do and sell to chip companies like intel. another is to mass-produce chips, that requires a very expensive unit. my bet is they rent a plant and organize the production.

      • SimonC

        I’d say the most logical plant to rent would be….Sony 😉

        Due to their long standing relationship, they would most certainly know the their production capabilities; hence, it would be the most safest bet to contract out your designs for production there.

        Who knows for sure. Maybe they do have their top secret fab somewhere.

  • brave new world

    Sensors and lens are linked – no question. If we all ask for new sensors, then it takes a major change in MPXL, sensitivity, noise and video read-out capabilities.

    If Nikon wants to keep the owners of lens happy and still wants to sell new lens, then they need to improve sensors step-wise. A jump from 12 to 24 or 24 to 34 MPXL will
    lead to a big frustration for all lens owners, who realize the lpr limits …

    Nick Kahn explained the “why” – the commercial aspect is: they will not make a leap, because it will frustrate too many of us.

    Just consider the Leica S2 – the new system required new lens having much higher lpr …

    • Nick Kahn

      brave new world: They may give us significantly higher res at moderate prices, if only to compete, and of course some will be able to take advantage of the increased res. For most of us though, and this has been said by many including Thom H. and others, there are a lot of other ways to improve our photography. I agree that the lenses need to keep up and of course some do. Nikon used to have perhaps more of a lens focus and if sensor/camera development stabilizes due to IQ levelling out, then maybe they will try and mass produce excellent quality lenses in higher volume at somewhat lower prices. Probably not, but I can dream.

  • Broxibear

    The one thing about this blog is that it does surprise you, I understand the “my camera’s better than your camera” posts, but “my sensor’s better than yours” ?…really?…what next, “the magnesium alloy on my dslr is 1mm thicker than the one on yours” ?
    For some announcements/rumours it might be a better idea to switch off the comments function.

  • testing new comments changes – now name and email are required

  • Dedpihto

    size…ISO it’s just nothing, IMO.
    if new camera’s own Nikon-RGB sensor it would have been fine even with a smaller size and working ISO

  • nikon has come a long way since the D2h sensor, a flop that likely taught them how to do it right. I’m looking forward what the next FX sensor (for D4) will be like. Nice time to own Nikkors.

  • nikon has come a long way since the D2h sensor, a flop that likely taught them how to do it right. I’m looking forward what the next FX sensor (for D4) will be like. Nice time to own Nikkors.

    For the past 4 years since I started following digital photog, only the D3’s sensor has gathered so much attention from Nikon users.

    • Marck Bis

      I wouldnt call the D2H sensor a flop, especially not in the case of the D2Hs. Some of my best pictures was made with that camera.
      It has a relatively low 4.1 megapixel resolution, but the D2Hs has a reputation for extremely high image quality and excellent detail rendition.
      The colours were just..fantastic.

  • brave new world

    well the conclusion is – not too much of a surprise: the best gear in the hand of a fool …

    technology advances, but image quality has not too much … if you just check popular photography sites: it’s great to see some good examples, but there is always a lot less good ones …

    for those in the discussion, who are concerned about quality: any camera is good enough, the question is, what is possible – all discussions about sensor, lens and post-processing are about: increasing chances to get it done technically. Any artistic measure is not depending on technology …

    back to the subject: new sensors will help – if you remember: Kodachrome and FP4 ruled the wy many of us were taking images … now we’ve got all kind of films and can even change in postprocessing … an advantage, if we just press the button at the right time …

    • It’s always been like that, you had great film and gear and quality differed by user, just same now. But to say that IQ has not potentially gone up is not correct: if you need high shutter speed in asituation that has low levels of luminosity, a 3200 or 6400 ISo capable camera is a big plus. And even at base ISo, the amount of shadow detail you can receover from a D3s file is so much more than in the earlier days of digital, in a level unthinkable with film.

  • DaveyJ

    @ Nick Kahn and NR in general: The information in this post is useful and important. I have speculated what this lens and resolution relationship is as I have often compared my large format prints to 35mm and admit that the large format prints are superior from a really good negative or transparency. Anyone using such gear also knows there are images that are crop failures (they just didn’t achieve what the gear could do). This situation has had me looking at what lens I do have and most will not hold up much beyond that certain point. One advantage any photographer has with 35mm is volume and speed. Once the camera (image capture gear)is set up properly….then this need to balance resolution and lens, etc. becomes far more apparent. Again, Thank You!

  • Phil Koenig

    Once again, I highly doubt that Nikon has a “secret fab” someplace. It’s really not even remotely likely that they would have something like that, or could actually keep it secret.

    The reason why this whole sensor supplier thing is important is because it has become one of the key differentiators between the products of different companies.

    Film cameras only had a limited impact on the ultimate image quality obtainable with them – film and processing was a huge component. So you could argue that there was only so far that a camera could impact that.

    Today, if you don’t have a competitive sensor, you don’t have a competitive camera, period. I’m pretty sure that Nikon doesn’t want to go through another long period of years when Canon embarrassed them by having digital SLRS with double the pixel count of Nikon’s products. (Or, for that matter, fast eye-linked, USM-powered focusing, when Nikon was struggling with slow, noisy screw-drive focus using a single, not particularly accurate focus point)

    There are various ways to achieve the sensor goals, many of which don’t require you to setup a chip foundry. A good OEM relationship can work just fine, especially if it helps the manufacturing partner. (either by giving them volume that they need to make that operation cost-effective, or perhaps licensing some of your in-house technology to them to use in their own sensors)

    I’m also a little bit worried about Nikon’s lens mount. They will never be able to compete on an equal footing with Canon’s EF-mount series in very fast lenses because the F-mount is just too narrow.

    I love the fact that the mount hasn’t changed since 1959. But sooner or later, they may have to bite the bullet the way that Canon did in 1987.

  • DaveyJ

    About ten years ago I conducted a series of tests with the assistance of a large custom lab and found that certainly when it came to print enlargements or even just comparing contact prints larger format such as 8×10 will make a sharper print even in a size like 3’x4′. I was always struck by the fact though that the expensive large format glass, Schneider, Rodenstock, Nikkor, Fujinon, etc. was not as sharp per a given unit than the best 35mms. We humans are not using an eye bigger than a fairly modest objective and of course we have the benefit of having two eyes which aids in having some sense of depth, etc. As to the benefits of medium and large format compared to say a nikon D3x I myself would suggest that each photographer at that high level (very few who would ever do that) better do the tests themselves. I came away from my personal search with the belief that large format and medium format just no longer could hold me hostage. I for sure think that you must decide what do I want, big prints or something that can be used in iPhoto on a big LED TV in a slide show. If it is big prints I still would be using large format. If it was big panoramic prints I would still own the Fuji GX 617. But for limited edition print books or HD TV on an LED TV I have switched back to 35mm and that evolution also led me to switch from film to DSLR. So as we head into an era with higher possible resolution possible from 35mm DSLR I myself would suggest FF 35mm DSLR. It is the path I have taken with landscape photography and when there is action, with people or wildlife, etc. the switch to 35mm DSLR (often in DX THERE) was the right choice. But if I lusted for really big PRINTS I would probably never had switched. Also here in printing I have owned some big scanners (still have them) and big printers but the big custom labs have better equipment so those big prints should be made for you I think. Fact is you will run out of walls really soon!

    • Just A Thought


      Printing large prints of same target, people could generally NOT tell difference between shots made with a P&S Canon G10 and those made using a Hasselblad with a Phase One P45 MF back.

      • Nick Kahn

        Just A Thought: I understand that people, for various reasons (to sell cameras mainly) occasionally like to make those kinds of comparisons, but again, as discussed above, don’t confuse pleasing results with actual resolution. There is simply no way that a 12MP P&S sized sensor and associated lenses can deliver the same resolution (and IQ for a given level of technology) on a print as larger formats with much higher pixel counts. And I should specify “under controlled conditions” where best technique is used for each. The print size would need to be large also to clearly show where larger formats shine.

        This is not to say that files cannot be uprezzed and printed and give pleasing results. We are talking strictly about resolution here and to some extent smooth tonal gradations.

        It is helpful to remember that as MP counts go up, it is increasingly harder to get meaningful differences in resolution however, since you have to double MP count just to get a 1.4X increase in linear resolution. So, if you have resolved those 5 lp/mm on a 16×24 print with a 24 MP camera, it would take a 48MP camera with the same lens resolving power (80lp/mm in this example) to resolve 7 lp/mm on the print. Or put a better way, since 7 lp/mm exceeds most people’s ability to see fine detail up close, the 48 MP camera could give a print that was 1.4X larger on each side, or about 22 x 34 vs 16×24 at 5 lp/mm. So, you can see the same resolution on the print but at a larger print size when using larger formats and most other things being equal. However, lens resolution generally is not as good for larger formats, so you probably could not print to 22 x34.

        The 12MP P&S file can only give about an 11X17 print at 5 lp/mm (about 240-250 dpi) and that also assumes that resolution of the P&S lenses are giving 80 lp/mm. Maybe some do, but I bet many don’t, so prints would have to be smaller for the same res.

        The basic premise that it is getting harder to see differences in resolution and perhaps IQ on a print with increases in MP count is to some extent valid, but right now, there are still signficant differences. To casual observers, maybe not, but to some artists and professionals working in the visual arts and commerical areas, the IQ differences are still worth the huge price differences we see. That is changing however, and if you wanted to buy say a Sony for 2K (I understand the reasons why someone may or may not), you could get most all of the D3X IQ at base for a fraction of the price, so prices are likely to come down with respect to IQ and res as a general trend. And sensor technology is changing – there are much more efficient backlit sensors now and who knows what else is in the works, and if the Bayer-less Foveon concept is made viable as far as IQ and the other issues it seems to have, then resolution increases significantly right there, since every color channel in our current Bayer sensors is only using a fraction of the total number of pixels.


        • Just A Thought

          The guy who did this is a recognized fine art Pro photog and he made larger enlargements. The larger prints were also reviewed by other Pro Photogs. Apparently it was difficult for the Pros to tell which camera was used to make which larger print. Real world test with real world results. BTW, they were landscape photos which also helped to reduce variables. Did you read the article?


          I can’t wait till they start using black silicon wafers to make camera sensors.

          • Nick Kahn

            Yes, I had seen that and read it awhile ago and I looked at some of the comparisons back then but not since your post. I will look again when I get a chance – working today. Can’t remember all the details like print sizes and test conditions. I still stand by my statements above as they should be supportable if comparisons are made as I described, but I do not doubt that quality differences are narrowing. If resolutions were very close, then a major factor could have been the resolution of the lenses used, which as I noted plays a big role in what you can get out of larger format systems. Not sure if that was tested in the LL article if I remember correctly. You could easily waste all the potential resolution of a larger system with a low res lens or due to vibration. I agree that real world tests are good but controlled, documented variables are important in any experimentation also.


            • Nick Kahn

              Just a Thought: OK, I went back and checked that comparison which I had seen before. Technical details seemed fine and more or less equal, although I have no idea what the lens res was for each case and that can be a big factor. It looks like they only compared 13×19″ prints from the Epson 3800 (nice), so that basically explains things. To restate a couple points:

              To begin with, the difference in linear resolution between 39Mp and 15Mp is about 60%, which is significant, but not huge (3 lp/mm would go to 4.8 lp/mp). It is not as much as the res difference when going from 35 mm film to 2.25 in film for instance.

              The G10 (15MP) can give 5 lp/mm (limit of human vision res) on a 13×19 in print at 240 dpi if the lens had high enough res. The Hassy (39MP) gives more res but you cannot see it on a 13×19 print because the res exceeds human visual limits ( again assuming the lens gives high enough res). If the prints were a lot bigger, you would clearly see the difference, as I had noted in the earlier discussion (assuming lens res for each system is similar).

              The article author says the same thing essentially. I have placed an excerpt from his conclusions below ( hope this is OK with LL – it is an old article from 2008).

              Excerpt from the LL article referenced above comparing the Canon G10 (15MP P&S) to the Hassy 39MP:
              “Please note that what I’m describing here is really not new when it comes to comparing high-end 35mm DSLRs to medium format systems. We’ve all done such comparisons for years, and know that the advantages of large sensors and MF systems are best seen in large prints and in critical applications. The only thing that’s different now is that instead of comparing an MF system with a DSLR I’m comparing it to a digicam, though a 15 megapixel one to be sure.

              Be aware as well that these comparisons fall down when prints over about 13X19″ are made. Once the output resolution drops below 200PPI the advantages of a 39 Megapixel sensor over a 15 Megapixels sensor become evident. And, even when smaller prints are made, cropping becomes an issue.

              Also, though on prints up to 13X19″ differences are almost impossible to see, on-screen at 100% one can fairly easily tell which files are from the G10. There are artifacts visible at the micro detail level and one can easily see other hints of what one is paying for.

              But, where the rubber meets the road (or more to the point where the ink hits the paper), in medium sized prints it’s been almost impossible for experienced photographers who I’ve shown these comparison prints to to tell the difference. Scary.

              One final comment. Landscape and nature shots are one thing – models in a studio with fabrics, delicate skin tones and other challenging subjects are likely to be quite another. Also, I have no idea how well these files might hold up to CMYK conversion. We therefore need to keep expectations within reasons.”

    • LED TV not exist. It is LCD with CCFL or LED backlit.

  • DolphLundgren

    I highly doubt it; if Nikon would actually start to produce their own sensors, they’d make a big thing out of it, surely their words wouldn’t be left at “developed by Nikon” …

  • Gråskinkan

    Does anyone else get depressed when there isn’t a new Nikonrumors post for a couple of days? I’ve started hitting the whiskey!

  • DaveyJ

    @Just A Thought: I concur that on many HDTV screens many camera images pull about the same. Prints to a certain size look pretty much the same. I was at a place surrounded by cameras from all over the world and saw results from many of those cameras. The Point and Shoot photographers all admitted that the 35mm cameras were for the most part WAY superior. Lets take a 3’x4′ print or a 3’x5′ from a Canon G10 and a Hasselblad…. the Hassy will generally make the G10 photo look pretty bad. I am very familiar with Luminous landscape and several of the Canadian pro photographers and what they are shooting with and recommend. Too much is made of cameras and not the subject and what the use is. I did sell my Hasselblads but it wasn’t because of my results with P&Ss!! Still a P&S is a lot easier to carry and much faster to use than a Hasselblad. Ultimately we each buy what we think we need. I am not in any way making a case for big prints. I have a few on the walls in my house and a number of other homes and offices and labs have some….but my big print days are over.

    • Just A Thought

      It would interesting to see a further series of similar tests made by Luminous Landscapes.

      Say Nikon D3100 vs Hasselblad with P45 back. Canon’s latest Rebel vs Hasselblad with P45 back. Sony’s evil camera vs Hasselblad with P45 back would be most interesting as there is no mirror flap. Maybe DPRview could add the same to their reviews.

      I agree with you in that a camera is just a tool. What kind of end result is required and where it might be used is very important. Most important of all, in my humble opinion, is the human behind the camera who will actually create the photo in their mind and press the shutter button or remote release button.

      Spending some time on DPReview, over the years, I’ve gotten to see amazing photography from low cost Canon Rebel owners. Some Pro wedding photogs, using top end 35mm gear, who post on DPReview should be paying their clients instead of the other way around. While other Pro wedding photogs post pics which are jaw dropping good. To me it proves that a camera is just a tool. Some know how to really use the tool, while others wear their high end gear like some expensive fashion accessory.

  • FakeKenRockwell

    Whether it is manufactured by nikon or not is not particularly important. The crux here is that nikon is not licensing a sony product or using the same sony production lines. This is probably strategically impactful for Sony than for nikon in the short term. No ff sensor order from nikon means Sony has higher marginally costs. Nikon can probably justify it with their premium market position. So it makes sense that D3x is in the 8000 usd range

    • Just A Thought

      The crux here is that nikon is not licensing a sony product or using the same sony production lines.”

      How would you know this? It was only stated that “image sensor from the Nikon D3100 is “developed by Nikon””.

      “So it makes sense that D3x is in the 8000 usd range”

      Does it make sense that Canon and Sony are able to sell cameras with high pixel count FX sensors for less than one third of the price which Nikon charges for a D3X and yet are still able to make a profit doing so?

      Nikon has only recently gotten into the black. It apparently takes Billions of Dollars to build a chip fab which can produce sensors. It also takes years to build. Who would lend an unprofitable business Billions of $ so that business could build a fab which said business had no prior experience in running? The financial contraction started in late 2007 – around the time that Nikon would have required the money to start to build a fab.

      Could they have purchased an already existing fab? And paid for it with what?

      There is one interesting possibility. In my opinion, if the Chinese Government had built a fab in China over the last couple of years and they are selling production time. This IMHO would make sense for the Chinese Government. Chips have many military uses and having a fab in China enhances Chinese security. It also makes sense in that the Nikon kit lenses are built in China. The new body will likely also be made in China. Producing the sensor in China reduces delivery problems and delays. It also keeps all the money in China as they would not have to pay for imported chips – so good for the Chinese Yuan. Manufacturers would benefit from cost savings on shipping and exchange rate differences. Nikon could lease such a facility to produce a few million sensors and also stop the rumors about Sony sensors in Nikon bodies. Win Win for all concerned – well maybe not for Nikon Thailand and the Sony fab – in my opinion of course.

  • Broxibear

    Here’s a Canon white paper on cmos sensors all you sensor geeks might like

    Personally I couldn’t care less about the sensor or who made it, as long it does what it’s supposed to do…but hey, each to their own?

  • Back to top