Nikon’s substitute DSLR production ahead of schedule

The latest later level report at the Rojana Industrial Park in Thailand

Nikon started substitute DSLR production ahead of schedule (via Bloomberg):

Nikon Corp. (7731 JT), a camera maker, gained 2.7 percent to 1,739 yen. The company, which shut its main plant in Thailand due to floods, will start substitute production of digital single-lens reflex cameras in the country at the end of November, nearly a month ahead of schedule, the Nikkei newspaper reported.

According to the last official statement from Nikon, production was expected to resume partially in January 2012 and return to normal level by the end of March 2012.

I am still not sure if the D800 will be announced next week with the SB-910.

This entry was posted in Nikon D800 and tagged . Bookmark the permalink. Trackbacks are closed, but you can post a comment.
  • lets hope this is the end of natural disasters for Nikon till 2020 because we are all tired of the waiting game 🙂

    • i’ve seen D800 today.
      picsies are real.
      but what shocked me is: it really does have 36MPix. No Sigma-like 3×12 or sort of that.

      • sade

        After NR post about exif of some D800 photo it became obvious that the sensor won’t be 3×12 Faveon type.
        do you have any information about ISO of D800? How does it compare against D700?

        • iamlucky13

          With a pixel size now effectively known to be roughly comparable to that of the D7000, I would expect similar performance to the D7000, which itself came pretty close to the D700 if I remember right.

          However, we’re talking per-pixel noise performance there. Across images viewed at comparable sizes, you should find the noise less noticeable.

          • sade

            Even though the pixel density of D7000 is the same as the pixel density of a 36mp full frame sensor, they do not necessarily have the same SNR. As far as I know, the size of the sensor might impact on SNR and the larger the sensor the lower the SNR will be.
            Moreover, per-pixel noise performance is somewhat vague. We do not have such a concept in physics. If you mean that the size of the noise grain in 36mp photo is smaller than that of a 16mp sensor, I should disagree with you. If both sensors are built using the same technology then both photos (in 100% view) in the same ISO setting must look very similar. That means D800 (in 100% view) should look much noisier than D700. (more than one stop noisier.)
            However, by downsampling and the same size print, D800 output must look similar or slightly better than D700 output, noise-wise.

            • Keith

              “As far as I know, the size of the sensor might impact on SNR and the larger the sensor the lower the SNR will be.”

              There’s no “might” about it – sensor size is what matters; pixel density is, to all intents and purposes, irrelevant.

            • PHB

              No, you are completely wrong there.

              It is rather obvious that the pixel density is determined by the sensor size and the number of pixels. Thus the sensor size will determine the pixel density for a given number of pixels.

              Larger sensors do not cause the camera to capture any more light or improve quality or any of the things you will read on this board. All that a bigger sensor does is to allow the camera to capture light from a wider area of the lens.

              Since the F-mount is an SLR design, wide angle lenses have to be retrofocus designs which means that a sensor the size of 35 mm film is optimal. There is no real advantage for any other type of lens. If you have a 200mm f/2 lens on DX the results will be practically indistinguishable from a 300 f2.8.

              Smaller pixels mean that the quality of information coming from each is reduced but for a given sensor size the total amount of information available rises and so a higher pixel density will always win until the quantum limit is reached. Note that this is even true of diffraction.

              There is a good reason that Nikon designed their EVIL camera the way that they did, they understand optics and most people blathering on this board do not.

          • NikoNiko

            Nikon are up and running again! Woohoo!
            Thats coz they weather-sealed their factory like they did the D3!


      • Q

        If you’ve actually seen it, I’d like some input on its size and weight. Smaller and lighter than the D700 or about the same?

        • bigger, heavier. Feels like D3s with chopped bottom part.

          • Any chance of sample images?

            Hey Admin, what’s your hesitation in sharing the samples you have?

            • BartyL

              Keeps us guessing. Keeps us coming back.

            • “Ill I Am Lyin'”

              Does this only seem fishy to me?

     doesn’t resolve, nor is it a registered domain in the whois database.

              I call bunk on this guy’s comments. What he says may be true (either because he’s telling the truth or because he’s extrapolating facts from NR’s posts) — but it’s obviously a pseudonym (or a fake handle altogether with the intent to mock). Take his comments with a grain.

            • it is of course fake to protect innocent.

              as long as i have no profit from it i will stay covered.
              sure take everything with grain of salt but as D3s owner i tell ya, i will pay 4k Euro for this baby IN A HEARTBEAT.
              it is awesomestest Nikon camera ever made and we can only thank competition for forcing nikon to do it instead of another minor update.
              It beats D700 in DR, noise and performance. Have ISO 100 as base, display is on sun better visible, and have FINAL resolution for this kind of camera. You cannot get more megapixels on 135. You will never need another camera except for replacement for faulty one. This is it.

    • To me the waiting is a very distant second compared to being concerned about the impact those disasters have upon those affected, including Nikon. Hopefully the people and families impacted will recover and the camera production as well.

      • Anonymous

        My thoughts exactly. Thank you for remember the multitudes of people who are suffering while we bitch about not having the latest camera.

      • Dchino

        Give it a rest, martyr.

        • Dchino,
          Your remarks says enough about you that I don’t have to, so no further comment.

          • Bill co

            Bob Ash,
            You want a different site if you want to send your regards to the families who were hit by the freak weather. This is a rumour site, about Nikon if you weren’t in the know….so let people fan the rumour flame and take your rather transparent attempt of pity else where mate. No will have to find another outlet for you an your care cup, coz WE only care about cameras and when we can have them so I can say to whinging do gooders like you – look at my new camera EVERYTHING else is irrelevant. Thanks for reading =)

    • fxed

      Maybe you should be tired of all the suffering and death caused by your annoyance!

      • Josh

        I agree. Don’t you love these ‘holier than thou’ morons that come on to a photography website and then wax eloquent about how we should all be thinking about the suffering in Thailand etc? As if we don’t! If these fools are so concerned about the suffering in Thailand, why don’t they fly there and help with relief? Of course they won’t do that. It’s very easy to sit behind a screen and chastise others for daring to want a new camera. Idiots of the first order, I say!

        • Art

          I agree. The best ways we can help those in Thailand are:

          1) Fly there and help
          2) Donate to a charity that has people in Thailand
          3) Buy a D400 as soon as it comes out

          Sitting around and whining accomplishes nothing.

      • Making such remarks about people, you folks are either looking in a mirror while you’re making those remarks, or failing to read notes correctly at all. Or both. Most likely both.

        Whining about cameras not coming out or about people expressing concern for the people impacted by disasters accomplishes a heck of a lot less than being concerned for the people impacted *alongside* of being interested in new announcements. And yes, I do donate to charities that help people in natural disasters like this one. Do any of you?

  • T.I.M

    Japanese people are strong and brave, same thing appening in France, it would take them years to recover !

    • cpm5280

      If it had happened in France, they’d have panicked and invited the Germans over to fix it for them.

    • burak

      ok you re french we got it, you could still put an h before “appening”
      pronounce it or not 🙂

      • T.I.M

        Me put an h before appening ?
        It’s never gona appen !

        • PeterO

          @burak (aka beet), We Canadians are used to hearing the Quebecois doing the same thing. We just don’t get why they put an “h” where it doesn’t belong, i.e. We don’t play ze ockey in New Horleans 🙂

          • Jesus_sti

            And want to you think we think about english who speak french.

            Et tu penses qu’on se dit quoi de la part des anglophones qui parle français. Se sont des langues secondes c’est pour ça que l’on fait des erreurs !


            • I always learned that even the French start names with a capital 🙂
              Ironic, really, erroneously writing about others makking errors…

            • Patrik

              Hey Freddy.
              Frenchies don’t capitalise “english” and “french”, nor do they capitalise days of the week, months, etc. Its just that way. Forgive him. I think its great that blogs like this are international. Peace man! (That’s all us Canadians want!)

            • PeterO

              Hi Jesus_ti,
              I regularly work with many French Canadians and we enjoy teasing each other about our pronunciation of the other’s language. On occasion, I also need to speak German, Italian , Ukrainian and Russian. We always give each other full marks for trying in the other’s language. We regular Canadians consider all the different regions as a beautiful tapestry and the Quebecois are a part of that.
              Now, back to our discussions about the D800 and SB910.

            • burak

              hey freddy,

              it s just that i dont like using capitals in the digital universe communications, except for TLAs (and job applications of course)

              you know it wasnt really critisizing right?

        • Discontinued

          @ T. I. M.,
          means you are never going to be happy. How sad for you.

    • Frenchie

      Japan is now located in Thailand… Hmm…
      New Orleans has returned to France??

      • Art

        Yea. We sold it to pay down the national debt. We only owe 15 trillion now.

        • 120-300 OS

          Acording to bloomberg it´s even 211 trillion a bit more don´t you think

          • Art

            Yea. The 15 trillion is the lower figure of what we now owe. The higher number includes our obligations that aren’t due yet but we have already spent. (Social Security, pensions, etc.). In short, we’re screwed and will be looking to sell Alaska back to the Russians shortly. The way the Mexican Cartels are running, soon they will have enough money to buy back TX, NM, AZ, and southern CA and we might just be desparate enough for cash to sell these states back to them.

            • Anonymus Maximus

              Alaska to the Russians and Mrs. Palin will candidate for Russian President?
              Might turn out to be the most brilliant move ever.

            • BartyL

              @ Anonymus Maximus

              What if she were to breed with Putin and produce a race of monster children? Think, man, think! She must be kept as far away from Vladimir as possible…

      • T.I.M

        And do you really think that the Nikon Thai factory is run by Thai managers or Japaneses one ?

  • Lev

    Does NR mind commenting about this DigitalRev post about the D800?—the-best-goes-to-those-who-wait-6474-article.html

    I figure this is another rumor that is worth mentioning!

    • VcBabis

      This post from DigitalRev is more than a year old…

      • Lev

        It was actually updated recently… nov. 23rd 2011

    • The Manatee

      Well, now someone is clearly wrong. You couldn’t get two more divergent rumored specs.

      • The Manatee

        I do hope DigitalRev is right. That would be the true D800 successor, not some 36MP FF.

        • Unless the D800 has pixel binning and/or sRAW modes that will allow lower resolution RAW files. I seriously doubt Nikon would release a 36MP D800 unless it out-performed the D700 in IQ and Signal-to-Noise. Nikon has NEVER took steps backward, you can bet they won’t start now. A lot has changed in the last 3 years.

          • sade

            Unless they introduce another D700 like full frame with lower mp and better image IQ. (which is unlikely).

          • Keith

            “Nikon has NEVER took steps backward, you can bet they won’t start now.”


            The D200/D80 sensor was a piece of utter crap compared to the sensor in (say) the D70…

            • Patrik

              Hmmm… I liked the D200. It was a huge improvement over the D70. Had the D200 for over 2 years before “upgrading” to the D300. Sure the D300 was better, but not by much. Cleaner at high ISO, and that screen!, but everything else were empty promises. Within 8 months I moved on. D700, now that was a change! Finally the focus “worked” and there is magic in the subtleties. Any body else liked the D200?

            • Yagion

              Really??? My experience from D70s to D200 wasn’t like that!!!

            • Andrew

              Keith, wrong!

              The D70 was a great camera, but the D200 was better!

              When released, the D200 had arguably the best color rendition of any digital camera and its exposure was superior than most any camera available at the time. Its LCD was awesome with beautiful color and detailed, sharp image. The pictures taken by with D200 is essentially identical to that of the D70, but the D200 has better ergonomics – its faster and more intuitive to use. Do your research.

          • PHB

            Nikon know what they are doing. If they are doing 36MP it is because they have found it to be better.

            People who are considering buying a $4000 camera can be expected to have a reasonable PC.

            The D800 will definitely have some sort of pixel binning scheme, even the D40 had that, you can choose output of JPEG at different resolutions.

            What you cannot do is to get RAW output from the sensor at anything other than the sensor because RAW is by definition the raw output from the sensor.

            That said, a 36MP JPEG will probably have higher quality than a binned 18MP file masquerading as RAW. But folk here would turn up their noses because its not ‘RAW’ – completely forgetting why RAW is better in the first place.

        • Keith

          Sounds just like a Canon 1Dx, in a smaller(?) body…

    • photo-Jack

      I really hope that this report is NOT right! Just the sentence “we believe Nikon has been focusing on – above all – boosting the video performance of the camera” is what I don’t wanna hear at all. I want the best of what a still camera can do today. And for a supposed price tag of $4,000 it is well rectified to expect this. If I would want video I’d buy me a dedicated video camera. If a video body with an F-mount would be available for a decent price that would be something I’d consider.

    • Now a complete set of new specs and it says 18 MP.

      NR Admin – What happened to the release date? It’s time to whip Canon finally!

  • Rami

    Please .. Please …please I wana buy a camera . I,m really tired of waiting . Or at least announce it “Nikon” it won’ t coast you a lot.

  • Dweeb

    I don’t get how you move high-tech equipment out of a flooded manufacturing facility and have it up and running in a month at another location. Maybe they’re just shipping parts to the Rebel factory somewhere.

  • lev

    Can NR please comment about the new digital rev rumor about the d800?!

    • I think they are wrong, 100% wrong. There will not be a 18MP D800. There will only be a version with the AA filter. The D4 may be 18MP, but not the D800.

      • BornOptimist

        I think you both could be right. Could very well be Nikons version of the Fuji S5 sensor. Both 18MP and 36MP in one sensor.

        • Admin posted metadata from a sample he received, purportedly from the D800. The pixel dimensions clearly indicate 36mp final output resolution, not merely pixel count (ie S5 or foveon).

          • BornOptimist

            The metadata shown by admin doesn’t say anything other than it was from a 36MP image. The S5 records images with size 4256 x 2848 (12.1MP), yet it is a 6+6 sensor, so in theory the D800 sensor could very well be a 18 + 18 sensor.

      • I liked the rumor that there will be TWO new FF cameras:
        D800 with 36Mp – 4000$
        D700s with 18Mp – 2600$
        Do you think it is possible?

        • Rob

          You won’t see a new full frame for $2600 the way the Yen has moved vs the dollar.

          • Mike

            You will, if you don’t live in the US. LOL

          • My question was not about prices.
            I’m interested in new D700-class FX camera with NORMAL megapixels, good video and two memory slots.

            • Rob

              Actually your question WAS about prices. You asked if a D700s for $2600 was possible. If you weren’t asking about price as part of the question, you wouldn’t have listed the prices.

            • Rob, you can see the word “TWO” written with all large letters.

        • FM2Fan

          Yes – but it it will burn many peoples budget

      • I’m an admitted NR addict, but I so hope DigitalRev is right on this one! If the D800 turns out to be a 36MP video-capable version of the D700 with no other ‘improvements’, I’ll be replacing my soon-to-die D700 with another D700.

        Hopefully we’ll know the difference between fact + fiction on the 30th 😉

  • Jabs

    Great news from Nikon. They moved their Manufacturing Facility to another location and DSLR Production resumes. Sony also resumed Production, so good news for everyone except maybe the people of Thailand.

    Here’s to the upcoming D800, D4, D4X and D400 in 2012.

    • No, there will no be a Nikon D4 and possibly not even a D400. There will be FF camera for masses in same body as a D7000, most probably it will be introduced around next Photokina.

      • sade

        There will be definitely a D4 in 2012 to compete against 1Dx and I bet it will 1Dx killer!

      • Richard

        If you mean a body to compete with the Canon 5D MK II (about to be MK III), Nikon needs to get one out this coming year, preferably in the first half.

    • I mean there will not be a D4x.

      • ISP

        +1 no D4x in 2012 that’s for shure !

      • Douglas Adams

        The FF sensor in D7000 body is probable just as much is probable that Hajduk Split wins the Croatian football championship! 😉

    • T.I.M

      D800, D700s, D900, D400 or D4 will be pro DSLR and made in Japan factories.

      • BornOptimist

        D300/D300s was not made in Japan, but in Thailand. Only D3/D3s/D3x/D700/F6 is made in Japan

      • Jabs


        You really believe all that? – lol.
        DX Pro bodies are not made in Japan and what do you now expect a D900 to be?

        A small bodied FX or what?

        I expect a D4, a D4X in FX and a D400 DX Pro body and don’t know about the rest as the Market is getting really weird due to financial crises all over.

  • been there guy

    So the tide has turned?

    I hope we will not be sold to a bunch of previously submerged camera parts without disclosure. The previous post about the deeply discounted P7000, I have my doubts about its parts origin.

    I know someone out there would disagree with me. But, corporation by its nature is greedy! If there is a profit to be made, they will take an advantage for it.

    There are too many corporate scandals examples out there, and there certainly will be more to come. I just hope Nikon is not among those corporations.

    God bless Nissan fan boys!

    • Jabs

      You really believe that there are any salvageable parts after so long a period under water and mud plus probable saline (salty) conditions?


      • PHB

        Yes, because the factory had quite a few days warning of the issue and at least some of the buildings are two stories tall.

        So any parts, finished goods and equipment they could store upstairs may be safe and 100% undamaged.

        My priority in that type of situation would be to save the equipment that was unique to lens and camera production. The production line equipment is more easily replaced.

        • Jabs

          It is more realistic to me that with such a warning, Nikon probably moved everything that they could salvage out of the Factory before it went under water. Most left there is probably water logged and rusted out plus contaminated, therefore Insurance loss to me.

          The items moved out will be now used to rebuild their Production lines in a newer location, hence the quick resumption of DSLR Production.

          • Richard

            There was a post by a member who lived in the area that there were several days notice during which time Nikon evacuated what they could in the time available.

            The risk to Nikon’s reputation of using flooded parts would be greater than I think the BOD could accept. Aside from that, the time and expense of examining and preparing parts would probably not make sense. Just scrap them, take the insurance and get the new production going as fast as possible. There is also the matter of the (Japanese) government loans to businesses affected by the floods which could help out during the transition period.

            If some of the production has been moved to Japan under the government’s temporary visa program, there will be an exodus of Thai workers in six months possibly resulting in another disruption of production. I just have to question how much Nikon would be able to move into Japan because of the energy shortage, if nothing else.

            If the production went to China, who knows what future plans Nikon might have.

          • Geoff

            They will be on ebay soon. LOL

      • @Jabs, I am no metal expert, but I doubt that a magnesium/aluminum camera body would rust to bits in a slightly saline water. As for the electronics however, that is a completely different story.
        If I was Nikon, and I couldn’t produce my PRO cameras reliably in a radioactive zone, I would take the production overseas to Thailand. It is the next logical location afterall. But then the worst thing that could happen, did happen. Their secondary site for production gets ruined by flooding. I suspect shortly after launch, say 6 months after, there will be a “flood” (pun intended, in bad taste I know, I hear the lambasting now) of refurbished PRO bodies on the market with the standard Nikon three month warranty.

        • Jabs

          Both Magnesium and Aluminum will rust if submerged hence why they have marine aluminum and such. Nikon would not sell things salvaged from the old plant, as in Precision Manufacturing pits in the metal parts would make them unusable and ill fitting, therefore more costly to deal with besides the ethical issues.

          Old metal will probably be melted and reused, but depends on its’ current condition. Most stuff is for salvage and therefore an Insurance loss. I believe that Nikon had enough time to perhaps move completed bodies or finished lenses out of that Factory as it is not like they did not see the flood coming. Air Cargo would have easily solved that and maybe China or Malaysia are good alternate Factories while Thailand is being rebuilt as most of the infrastructure of component Manufacturers are probably still in Thailand, so moving is of no value unless they too move.

        • iamlucky13

          Often manufacturers will not ship parts that have been exposed to even potentially damaging conditions. The liability of having large scale failures is simply too big.

          Several years ago, a car transport ship was towed into Portland, OR after having a ballast issue that left it partially capsized, without power (No ventilation in a humid, salty air environment), and out of control for several weeks. It had several thousand Mazda’s on board.

          Every single one of them was scrapped, because Mazda couldn’t be sure there would be small, but critical defects resulting. They were all insured, of course, but something like an airbag not deploying due to corroded electrical contacts could have led to a lawsuit that would have wiped out the value of those cars.

          Any inventory that Nikon could secure from the water is probably usable. They have a reasonable idea what conditions it was exposed to. Inventory that was exposed may or may not be scrapped. Some plastics, for example, can absorb water and expand, no longer fitting and having reduced strength.

          Things like metal bodies can potentially have a few copies sacrificed for metallurgical testing to determine if they’ve corroded too much or can be cleaned and used. Similar for things like glass blanks or possibly even cut lenses.

          I’d expect electronic components, motors, and complete cameras to be scrapped. Sometimes these can be cleaned up and be perfectly functional, but that has its own cost, and like Mazda, Nikon has to consider how much hidden failures can hurt their brand image compared to how much their future insurance premiums might be saved if they reduce their insurance claim that way.

          On the other hand, complete cameras and lenses are the most valuable things in the factory other than the manufacturing equipment, and much more portable. Those would be the most likely items to be moved.

    • “I hope we will not be sold to a bunch of previously submerged camera parts without disclosure.”

      Duh. Go back to your hole. Sony morons.

      I fully expect to buy my products from companies who UNABASHEDLY INTEND TO PROFIT from their productivity. If they say they don’t want profits, they’re:

      A) Lying,
      B) Corrupt, or
      C) Just plain stupid

  • elph

    The one thing that really caught my eye, bodies of water from 20-25cm deep inside a building are considered puddles?!

    • vinman

      It’s all relative – once you’re accustomed to canoeing to the market (now a wash basin) and swimming up to your second story flat, 25cm really does seem insignificant!

    • Art

      Yea. They are puddles compared to being in the middle of a lake.

    • uneven flooring.

  • FM2Fan

    Great – the crisis mgmt seems to be really exceptional. consider this being just an additional value of the brand

  • JonMcG

    Is it just me or is anybody else out there scared Nikon is going to really screw up everything that made the D700 such a great camera (but needing an update)???

    Will they be able to achieve same or better low light capabilities as the D700 with 36mp?
    We already know the cam won’t shoot as fast and that’s a huge negative for wildlife and sports shooters.

    I’m really crossing my fingers they get this right and the 36mp turns out to be something that I only need to worry about as far as storage space is concerned..

    • Jabs

      Perhaps Nikon has several upcoming cameras geared towards different segments of the Market including a proper replacement for the D700, as I don’t see this as a response to that body.

      I see the D800 as a response to an upcoming Canon 5d MK3 maybe!

      Just wait.

      • Richard

        I don’t see how this could be a competitor to the 5D MK III at the rumored price (roughly $4k U.S.).

    • jorg

      that´s just you.

    • @JonMcG – That is a very valid point, but I bet if you asked everyone who has been a long time Nikon shooter, you will discover that Nikon only gets better. I can’t think of a Nikon camera which came out in the last 5 years as being worse than its predecessor. With 36MP just arround the corner, I can only advise that people start investing in the home NAS manufacturers; those photos got to be stored someplace.

      • This is precisely why I recently published a blog entry about our ever growing storage needs, especially with a 36mp D800 on the horizon. A NAS is a must!

        • +1000 @ CTLG! Great blog bost on NAS, BTW. As an IT professional who works with PETABYTES of storage using industrial SANs from EMC, I couldn’t agree more about your assertions. A great website I recently found may help others find the perfect NAS device to fit their needs and budget.
          It covers devices from most of the major players in the SAN buisness. The finder tool recently helped me pick out a little gym of a device for my mother, a non technical person who needed a safe place to keep the family archive of photos as she digitizes the old photographs and negatives.

          • Richard

            Even if people are not using NAS, the need for fast connections to external storage has never been more evident. I don’t know whether the new Thunderbolt will find wide acceptance in time or whether USB 3 will suffice, but something fast , perhaps external PCIe (which is what Thunderbolt is based upon), is in order. It also needs to be easy and fast to backup.

            • Totally agree, Richard. Fast transfer rates are absolutely needed, more now than ever! The future holds many secrets, let’s see how this all pans out.

          • Great link Dr SCSI! This will be very helpful for people looking for the right NAS. I might add this link to my blog, actually. I will of course, give you credit for sharing the link. 🙂

          • photo-Jack

            Can you tell a computer dummie why the heck one would need NAS (don’t even know what that is) compared to some Terrabyte-harddiscs just for the photos?

            • Did you read my blog entry about NAS? It will explain most of your questions.


            • Photo-Jack

              @ CaryTheLabelGuy

              Couldn’t reply directly on your post anymore therefor this way…

              Thanks for your article. It helped to some degree. As far as I understood RAID it partions the data of one file into 4 pieces (given its RAID 4) and if one HD fails RAID is able to reconstruct the complete data out of the remaining 3 HDs.
              So far so good. Given I’d use 4 HDs 1 TB each in RAID 4 how much storage capacity would I have altogether? 4TB or just 1 TB?
              …and why do you recommend RAID 5 (That’s 5 HDs right?)

              Unfortunately the Thai Flood didn’t only bear down to the Photo industry but also prices for HDs are skyrocketing. Thus I suspect prices not to be $200-700 but $700-1500 for such a NAS equipment …?
              And by the way, what brand of HDs do you take for the NAS? Are WD a good choice?

              Since you seem to be a storage expert I’d like to have another question. Being just busy with a back-up of my photos, I realized that a few pics were blackened by about one third of the photo-size after copying them from one to another HD. I have no clue of what could have happened. Do you by chance?

            • Jabs

              A solution for fast storage is in several variants!
              For speed of transfer, you use Raid-0 and for redundancy to have your files available still if one hard drive fails, you use the higher level of Raid with multiple drives. Sometimes this is overkill and it often is simpler to just buy one fast SSD (solid state drive) and use as your man drive where you put your OS and then get some plain multiple terabyte drives for additional storage. This of course applies to desktops but some laptops have two or more drives.

              Windows 7 and Vista are known to be really slow with file transfers as they check them first and thus an agonizing wait. I use TeraCopy to solve that. Win XP is fast in file transfer but I also use that on XP. In Linux, the file transfer is fast and so is OS-X.


              File transfer is a product of how fast your drive is in transfer speed and thus use the fastest you can – Raid-0 and if you can afford it, get a Raid-0 SSD drive setup and that to me is faster than a NAS, but the NAS requires less thought and capability to set up. Get some automatic backup software and backup to an external drive often, as even Raid arrays fail.

              Thunderbolt and USB 3 are faster but not many cards available yet, but some laptops and desktops now have that option. An SSD drive with a Thunderbolt or USB 3 connection would be the fastest route to go or you can use a PCI-E SSD which is faster but also a lot more expensive. LaCie makes some new and really fast drives that are perfect for Macs with the Thunderbolt connection.

            • Jabs


              Four 1 terabyte drives in a Raid array would get you a 1 terabyte Raid array and not four terabytes of storage, roughly.

            • Richard


              Assuming you had a 4 drive RAID 0 controller (the most common ones are two drives) you would have a 4 TB (4×1 TB) RAID Zero array. In any RAID 0 array failure of any drive in the array results in the loss of all data. It is the highest risk RAID configuration. It’s benefit is speed. If you want to maximize speed in a rotating drive RAID 0 array, partition the drives to use only the outer two thirds or so of the drives. A RAID 5 array consists of a minimum of three drives. Two contain data and the third contains parity data. The array can recover from the loss of any one drive, but not two. A RAID 6 array is basically a RAID 5 array with a hot spare. It can survive the loss of two drives. A RAID 1 array is a mirror which can reduce down time in the event of the loss of the primary drive, but is subject to other problems. If there is a corruption of the system or data on the primary drive, it more or less instaneously happens to the mirrored drive and the system and/or data may be lost. I won’t bother to go into a RAID 10 array.

              NO RAID ARRAY IS A SUBSTITUTE FOR A CURRENT BACKUP! There are too many things that can happen to go into, except to say that the only thing of value in your computer is the data.

              At the present time, SSDs are probably too expensive to store large quantities of data on except at the enterprise level. The most common combination outside of the enterprise is a SSD boot and/or applications drive and a rotating data drive (which is frequently a RAID Zero array both for size and speed). There are some who have used SSD RAID Zero arrays for their boot drives and/or PS scratch dive.

              As you observe, PCIe SSDs are largely the realm of enterprise systems at the present time. I would not be surprised if Dr SCSI could give examples of real world benefits of such systems. Let’s hope the cost of them does come down as they do outperform SATA drives in tests.


            • PhotoJack,

              Richard’s explanation is very well done and presents both a description and also the pros and cons of each RAID flavor well.

              NAS is better when you have your data striped (distributed) over 2 or more drives. The reason is that your internal drive is very slow compared to your RAM. VERY slow. It’s the slowest component of your computer (besides typing on your keyboard 🙂 ). It becomes very painful with really big files, they take a long time to pen and a long time to save.

              Distributing the files across 2 drives in RAID-0 means 2 drives are working on the file at the same time. Makes it about twice as fast to open or save a file. The more distribution, the faster the files can be opened or saved.

              SSDs (solid-state drives) are even faster but they’re smaller and more expensive right now.


              PCIe SSDs have the same benefits as internal SSDs but you’re right, they’re expensive. Any SSDs are expensive still right now. Plus they’re really small compared to most hard drives. Right now SSDs are best used for the operating system and for large working files like PSDs or TIFs that you want to open quickly and save quickly. Then move them to slower storage where you have more space.


              RAID-5 is actually really good for read and write speeds if you have >3 drives. Also, with more drives you don’t have to keep all your parity information on one drive if you have intelligent parity distribution (i.e. no disk’s parity information is on stored on itself).

            • Photo Jack,

              I personally use all Hitachi drives. I’ve found them to have the smallest fail-rate for me. I’ve got some Hitachi drives that are over 6 years old that are still kicking it and are still healthy. But honestly, it doesn’t really matter. A HDD can fail at any point, no matter who makes it. They’re all about the same, honestly. Especially this day-and-age. Quality has really gone down in HDD manufacturing over the last few years.

              Either way, if you guys are going to be buying drives, do so NOW! The prices are only going to go up from the crisis in Thailand.

            • Richard


              I have been well pleased with Hitachi drives as well. They are also backwards compatible with more SATA I controllers than some other drives.

              The reason for my posting though is that I have discovered that the way a vendor packs drives has as much or more to do with their mortality than the construction of the drive. One vendor I quit using some time ago did such a poor job of packing drives, even those that passed being zeroed before being placed in service, that many failed within the first 30 – 90 days of being placed in service. When poorly packed and placed in the loving hands of UPS and being pounded many of the drives simply were destroyed. There was an article some time back by a European data recovery service that a large percentage of the drives sent to them to attempt recovery on had bent spindles. A bent spindle mean that the drive will die much sooner than later. Despite all the quotes of how many static Gs the drives are designed to survive, experience indicates that these figures are engineering calculations not supported by actual experience. Interestingly, one of the major design differences between the WD Blue and Black drives is that the Black drives have bearings at both ends of the spindle which helps some, but is certainly not an immunization against physical abuse of a drive. Notwithstanding what the engineers may say, it is my opinion that rotating drives continue to be delicate devices which should be treated with care, including putting them on a smart (voltage regulated) UPS to prevent crashes due to poor electric supplies and outages. Voltage fluctuations do nothing good for a computer. (No, I don’t work for or have a financial interest in a UPS manufacturer. I just think they should be considered an essential part of a computing system.)


            • Jabs


              If you are using four drives, then you obviously are not using a RAID-0 array – hence my reply.

            • photo-Jack

              Wow so many good comments. Thanks a lot guys! The horizon did get quite a bit clearer for me.
              Just deceided to get me a new i7 PC with WIN 7 a SSD as boot & program drive and an internal WD blue 1 TB. Furthermore I have external drives like a LaCie 1TB / USB 3.0 to mirror the internal drive manually (as back-up), which to me has the advantage that my own mistakes as well as electronic mail-function is NOT instantly transferred to the mirror.

              However, I would like to come back to the (at least to me) strange phenomenon, that maybe 3 to 4 jpg files copied from a fully operational drive (no bad sectors) to another had this blackened areas (a right angle in size of approx a third of the jpg. Does anybody know something about this phenomenon?

            • Richard


              If you want to build a new computer, find someone locally who can give you a hand selecting components or try some of the web sites like where you can ask a lot of questions about specific hardware.

              Choices are, of course, determined by budget. If this is to be a Photoshop machine, the clock speed of the CPU has more to do with performance than having a gazillion cores according to a number of sources including Lloyd Chambers at his Mac Performance Guide web site (hardware test results generally apply to Windows machines as well).

              Get a motherboard that takes a lot of RAM 16 GB (4 x 4GB) seems to be a good performance/price point at this time.

              There are a number of SSDs coming on the market that support TRIM which is useful. (It’s pretty technical, but helps the drive live longer and work better.)

              If you budget permits, get a small SSD to use as a scratch disk for PS & Lightroom. Some people have used a “stacker” PCIe card to install RAM on to use as (RAM) scratch disk because it is a faster connection than the SATA connection.

              I am not really a fan of the WD Blue drives, but I suppose they are OK. Like CaryTheLabelGuy I like the Hitachi drives, but that is just my experience.

              If you really want to run wild, you can look into various RAID solutions, but that starts running the price up in a hurry.

              You might try posting on the NR Forums where it can be discussed in more detail.

            • photo-Jack

              Thanks again Richard.
              I was actually researching for at least 2 months what kind of system to buy as photoshop machine. Most of the computer-freaks turned out to be good advisors for gamers and the photo-folks I’ve been in touch with are not deep enough in computer technology. Thus all my system-questions like “whether AMD X6 or intel i5 or i7” “Why big RAM while having a fast SSD” “Which graphic-board and is Grafic-board RAM an issue for Photoshop applications” etc. remained unanswered.
              Thus I did the best of my knowledge: i7-2600K with 4×3,4 GHz, Corsair Series 3 120 GB SSD, Nvidea Geforce GTX 560 1024 DDR5 RAM.
              Concerning the HD I had the choice between the WD blue, a Samsung and a Seagate Baracuda Green while the later apparently does have the worst image. (Hitachi wasn’t offered)
              Also concerning displays the answers were whether pointing to very high prized Eizo or were Zero. Thus finally I decided for a 24″ ASUS PA 246Q. Plus a Quato iColor Display with an DTP94 for calibration. Already went over my budget limit since PS CS5 & Lightroom prices are also not cheap. Will see how this works out.

        • Careful with some of those older NAS drives. The old ones were pretty much all just under powered embedded linux boxes with poor transfer rates. The one you mention in your article (ReadyNas RND4410 has at best transfer rates around 30 MB/s for read and about half of that for writes … about 5 times slower than what you get from a good internal harddisk (15 times slower than what you get from a good recent SSD drive)!

          Modern NAS boxes have finally caught up (anything powered by an ATOM processor or better should be fine) and give you transfer rates >100MB/s for read and write … If you are shopping for NAS Boxes then make sure you google for speed benchmarks before buying a box! Moving big files around can become a PITA with the wrong gear!

          • Hi Stefan,

            Speed is deffinatly an issue when it comes to NAS boxes. I actually upgraded my RAM on my ReadyNAS NV+ and it helped transfer rates drastically (especially write speeds) and only cost me $25. Even with the slower NAS boxes, speed is still fast enough to do most things with ease. Large file transfers can take a little longer, but not a HUGE issue imho. Remember, you’re not running your PC or Mac off the NAS – it’s only there for file storage as well as serving those files. Even a slower NAS is much better than having no redundant file storage system at all. Remember, the faster boxes usually cost a lot more. Not everybody has $1400 in their budget for the fastest NAS – especially when a $650 4x1tb has 75-85% the performance of the more expensive boxes.

            I run my OS on 4 x 80gb Hitachi SK80s in RAID 0. And perform nightly OS backups to my NAS. My NAS stores and serves all of my files (photo files, media, movies, music…etc). My NAS performs daily backups to an external HD as well as backups to a popular online backup service. Every month I move all of my photo files to DVD as well as another external HDD and clear the space on my NAS; This keeps my 2.75TB of usable space (4x1TB RAID 5) on my NAS fresh.

            This seems to work pretty decent for me and provides several levels of redundency.

            • Jabs


              Actually, the problem with drives now is capacity as in their denseness required to get a certain gigabyte or terabyte capacity.

              More platters has meant less reliability and thus the smaller drives still outperform larger drives due to heat issues. Same package with more platters crammed in there means heat goes up and so does failure rates. Best drives are usually SAS Enterprise drives or Solid State drives plus you should use a drive cooler and then get longer reliability.

            • @Jabs, Although there are more platters in a modern HDD, they are not indicative of failure rates (measured in Meantime Between Failure MTBF). In recent years, perpendicular alignment of the magnitites on disk have provided the exponential gains in aerial densities; not so much the one or two extra platters. You are however right about heat being the killer and SAS Enterprise drives being more reliable. I will disagree about SSD and their relieability factor for general computing use; but in other applications SSD may be more appropriate because of environmental conditions such as vibrations, or lack of gravity (such as in space). Going back to heat, it is a function of the raw wattage used to spin and operate the drive. I have seen 15K RPM Seagate Cheatah drives get hot enough to litterally cook on, despite active cooling on the hard drives. So, here are my recommendations about buying a drive/s which will typically last longer, but definitely not a speed king.
              1.) Find a manufacturer who has been making hard drives for a while and has a good reputation.
              2.) Look for a drive that spins at 7200 RPM or less, preferably 5400 RPM. Higher RPMs require more energy which equals more heat.
              3.) Look for a capacity which has the highest aerial desities (i.e. MegaBytes per square centimeter).
              4.) Go GREEN! The environmental friendly drives typcially satisfy 2 and 3 above, while requiring less energy in both the active and passive modes of operation.
              5.) As Jabs states, provide good ventallation for your drives, possibly active cooling by fans.
              Remember, your not trying to get the best performance out of your drives that you use to store your photo library on. It is more about reliability. If you want both, then look to an internal PCIe RAID controller card for SAS/SATA from the likes of LSI or Adaptec.

          • @All, I have read many of the comments here about various storage technologies and their implementations. Many of the remarks were close to the bullseye, while some missed the target all together. As someone who makes his living in the IT field, I would like to add the following to inform or possibly help those looking into a NAS device to protect their artistic works.
            1.) Technology is a means to an end. Storage is much like photography, we have many options and many methods to achieve an end goal. In the storage world, our goals typically lie somewhere between highly available or lightening quick.
            2.) In IT, the main goals all seem to revolve around High Availability (HA) and Disaster Recovery (DR). A modern NAS device can provide you both. So I will try to stick to those two goals with a photographer’s mindset. The simplest NAS should give you a minimum of HA by leveraging RAID 1 (Mirroring). This ensures that if a disk fails, the mirrored disk can take over without interuption or data loss. In terms of capacity, RAID 1 is an expensive option, because you only get to use 1/2 of your disks in the mirror for storage, the other 1/2 is for duplicating the first half. It is also the easiest to implement at both a hardware and software level.
            3.) RAID 0 (Striping) has its place for performance. In its simplest form with just two disks, it splits your bytes across the two drives for reads and writes. It is also twice as likely to fail as a single disk! I wouldn’t recommend it for anyone attempting to achieve the goal of highly available.
            4.) RAID 5 provides a balance between performance (RAID 0) and high availability (RAID 1) by using 3 or more disks in a striping fashion which also have a check bit, known as parity. RAID 5 is more effecient in terms of capacity, but yet provides the much needed goal of high availability. Without going into the grueling details of how parity works, I will provide you just enough so that you understand it at a higher level. Basically, in a three disk RAID 5 array, if one of the three disks fail, there is enough information on the other two disks to reconstruct the missing bytes from the failed disk. You also have the benefit of being able to continue to read/write data, albeit much slower. If a single disk fails, your RAID 5 is running in a degraded state, and there will be extra overhead at the software/hardware level to perform disk input/output operations, until you replace the failed disk and the array fixes itself. If two disks fail, you lose all of your data, hence RAID 6 was born.
            5.) RAID 6 is an expansion on the RAID 5 idea, but it tolerates two simultaneous disk failures; a minimum of 4 disks will be required.
            6.) Additionally, there are many other forms of RAID, including RAID 50, RAID 60, etc. The important thing to remember is to keep your personal goals in sight. What is it that you want to achieve? Most NAS solutions are designed with high availability in mind, and speed second. They are not really designed to replace your primary boot drive in your PC (although in theory, they could, albeit much slower). I know someone will probably bring up the topic of SAN, but for home use, at this point in time, I would say forget it; it is way overpriced and WAY over the heads of your typcial user.
            7.) I saw many discussing the merits of SSD for sheer speed. Which is fine if you have the budget for your bootable drive, but not so fine for archival of your photographs, unless of course your name is Bill Gates and money is no option. In my company, we paid 1/2 million dollars for 6TB of Enterprise SSD for an EMC VMAX SAN. From that 6TB, we only get to use 3TB, as the other three are part of mirrored strategy at another data center. Yes it is fast, and it is typcially only used for real time data aquisition projects, or to make SQL database queries REALLY quick.
            8.) Back to reality…I said earlier that a modern NAS will provide both HA and DR. I discussed how it provides HA by leveraging RAID technologies, but I didn’t tell you how it provides DR. If your house burns down (God forbid), your home NAS probably won’t survive and your data will be lost. However, if your NAS supports backing up to the “Cloud”, an offsite storage provider, your files should be nice and safe somewhere and retrievable after the insurance company has helped you put your home back together. The backup to a offsite storage provider is your disaster recovery ACE in the hole.
            9) Summary…take your budget into consideration and buy a NAS solution which fits your goals. For about $300, I picked up an older, slower, 2TB LG NAS device for my mother. It provided her, 1TB of useable, RAID 1 Storage and a built in Blue-Ray burner for backups. My mom is paranoid about putting stuff on the internet, so the “cloud backup” feature was a must NOT. Of the devices I looked at, I would say that the bang for your buck and features is with QNAP or Synology. I am certain their are other competitors products which are equally good, but those two companies stand out above the rest as they not only target the small home user, but they cater to medium businesses as well. Do your homework before you buy, but more importantly, identify your NEEDS before you start to shop. Keep your goals and budget in mind.

            • @Dr SCSI

              +100,000,000 BINGO!


              You should trust a dude named Dr. SCSI, especially when we’re talking about IT tech.

            • HAHA.

              “IT tech” is obviously a redundant term.

            • Richard

              @Dr SCSI


              A while back I discovered one rather specialized use for SSDs. It seems that the air cushion the heads of a rotating drive “fly” on gets too thin at about 10,000 ft MSL ((above) mean sea level) and head crashes are a real risk. This is not a problem on commercial jets where the cabin pressure is maintained at about 8,000 ft MSL. Not a whole lot of people will find themselves in this situation.


            • Jabs

              Some great discussions here and thus we all benefit. Nice to see this here instead of the usual – Nikon is lame and get rid of the Nikon 1 System.


              Everyone seems to gain from intelligent and rational discussions too.

          • @Stefan, most of the NAS solutions on the market run some variant of embedded Linux. Linux is typically better than Microsoft when it comes to tuned kernel which has been optimized for I/O. So Linux isn’t a bad thing, and the free factor keeps prices down. The biggest bottleneck for these types of devices is in the fact that they are Network Attached Storage. What I mean by that, is they use your slow antiquated Ethernet framing network to transfer files. Additionally, they will quite often have just one Ethernet port, the newer ones are coming with two or four for link aggregation, thereby doubling or quadruppling the bandwidth. You are right about the ATOM processor as being a good indicator that you probably won’t get a turtle, and moving big files can be a PITA, and do your homework. I will add one to that….make sure the device you buy supports JUMBO frames for Ethernet and be sure to enable it; it will help large file transfers, such as photos, but slow smaller file transfers, such as little word documents or notepad files. I will say though that CTLG does make very strong points about his purchasing decision. In the end, everyone needs to get whats right for them; and sometimes an older NAS device with disks comes in at bargain prices many can afford. Me personally, I have more money than common sense and I will probably build my own SAN!

  • Jabs


    Maybe this – Nikon’s substitute DSLR production ahead of schedule

    Should be – Nikon’s substitute DSLR production is ahead of schedule

    Clearer to me!

    Thanks again for your hard work and looking forward to the images from the upcoming D800 plus its’ release soon.

    • Steve Starr

      Actually, I was confused by the header as I read it to be some “Subsitute DSLR” production (like maybe a D710?) and not referencing a “Location change” for their production facilities.

      Maybe “Nikon’s DSLR production is ahead of schedule due to a manufacturing location change.”

      • Jabs

        Maybe, but a little awkward to me.

    • Rob

      No. You do NOT put “is” in a headline. See #4 here:

      • PeterO

        This is a situation up with which I shall not put.

        • Rob

          That’s actually not a rule in English. In Latin, you can’t end a clause with a preposition, but you certainly can in English. I’m not sure why some people were taught that, but it’s pretty silly. And even if it were, almost any sentence can be made to look awkward by arranging it incorrectly and using archaic words (see Yoda):

          I cannot put up with this situation.

          • PeterO

            Thank you Rob, obviously the meaning of my quote was not understood by those who continued to debate the grammatical point and who should have a quick look at Wiki Hypercorrection.

      • Jabs

        Much ado about nothing!

        • Rob

          You mean your post? Since you were the one who decided to comment on it, even being wrong?

        • Jabs

          It’s Shakespeare – clueless person.

    • Jabber

      I was really enjoying Jabs time away from here.

      • Jabs

        So was I too – that is, enjoying you not being here – LOL!

        • Jabber

          Maybe this – So was I too

          Should be – So was I


          Should be – I was too

          *rolls eyes*

        • Jabs

          Maybe you need to learn English composition and not focus on me – LOL!

          Plus, there was nothing that sparked my interest or needed my comments, as most of the ‘intellectually challenged’ posters had vacated the place including you.

  • FX DX

    I dont know how real are D800 prices. I don’t think that it will quite work if Nikon puts it at ~ $4K,but if it’s under $2K, it will totally change the game. Let’s hope for the best.

    • @FX DX, with any new launch, you have launch prices. As Nikon needs to make as much profit as they possibly can, they are going to maximize those launch prices and see how well they are received. Those who MUST have a new camera, will buy them regardless of the initial price. Those who just WANT the new camera, will most likely wait until the flury of reviews and prices have settled. (Un)Fortunately, I don’t make a living with my photography, so I wait until the refurbs hit the market; my current D700 has another 100K exposures to go before it needs to be rebuilt. But for me, the D800 is less interesting than the D4 which is coming up. BTW, if you want the upcoming D800 for under $2K, you might be able to get it in 2015 for that price. But on launch, expect to pay slightly higher than MSRP for first dibs, and MSRP for at least 6 months after, and maybe a 10-15% drop a year from then. The price fluctuations will probably have more in common with the exchange rates anyway.

      • FX DX

        I just bought a used D700 in a good condition for a little over $2000. Considering that D700 is selling new for the same price for as it was selling 3 years ago makes me think that D800 will not go down much in price even a year from now. Nonetheless if video is superb, price is reasonable and IQ is better than D700, I will trade my D700 for a D800. Like you, I don’t make a living with my photography, so if it is close to $4K in price, I will certainly wait 3 years for it to come down in price. If it is close to $2600 (current D700 price), I will pull the trigger right away.

        At some point FX sensors will come down in prices and become more affordable. It’s just a matter of time.

        • Richard

          @FX DX

          “At some point FX sensors will come down in prices and become more affordable. It’s just a matter of time.”

          Unfortunately, it appears that getting the price of FX sensors down to whatever “affordable” may be will require expensive new equipment or a completely different technology from what we have known.

          Take a look at this article for a more detailed explanation.

          Thom Hogan has more recently estimated the cost of a DX sensor at $50 and an FX sensor at $500.

          I continue to be puzzled why Nikon have not made a larger than DX sensor similar to the APS-H sensor which Canon have made. Canon supposedly came up with the dimensions of the APS-H sensor as being the largest size sensor which could be made without “stitching” multiple segments of the sensor together. Most of the image sensors are reportedly manufactured using 200mm wafers with steppers that were sized quite some time ago. There is speculation that there are 300mm fabs being added, but no information whether the size of the steppers being used with the 300mm wafers have been increased.

          The 2006 article linked above expressed a belief that the equipment necessary to manufacture full frame sensors without stitching would far exceed the worldwide demand for full frame sensors. Whether things have changed enough since then to cause a different result is unknown.

          I do note that Canon appears to have abandoned the APS-H sensor size with their 1D X announcement which makes me wonder about their production capability, although the price of the 1DX is quite staggering.

          A critical issue in all of this is the ability to increase the yields of FX sensor production.


          • Calibrator

            I can’t believe $500 for the sensor in the D700.
            Maybe D3x, but not D700.

            • Art

              It has a lot to do with the price of the silicon wafers and erching. It doesn’t much matter what you do with the wafer but the wafers themselves are expensive as is the etching process. The larger sensors use more wafer and processing.

            • Richard


              I have no independent knowledge of the price of sensors, but Thom Hogan probably has as good information as anyone who is at liberty to say.

              Art has the basic idea correct, although it is not simply a matter of the cost of the silicon of the sensor itself. As the article, and other sources, says, the yields for a sensor go down as the size goes up. It has been suggested that the process (which, frankly, is rather old) presently used may result in yields of FX sensors may be as low as one (1) per 200 mm wafer. Take a look at the article and you will notice that both the author and Thom Hogan have a ratio of between 10 to 1 and 11 to 1 when comparing costs of FX sensors to DX sensors.

              Something needs to improve before the costs will come down a lot. Sony are reported to have committed to increase their sensor production and there is a new fan reported to be coming online. We shall see.

          • OsoSolitario

            Why wafers must be round? Is not possible to make them square or right-angled?

            • Pegdrgr

              As soon as you see a square or otherwise angular icicle you will have a shot at getting a square or angular wafer. Wafers are cut out of ingots or boules that are made by taking a seed of the semiconductor and dipping it into a molten pool of the same semiconductor. The ingot is slowly rotated and lifted out of the pool allowing the molten material to solidify and attach to to the seed in a precise molecular structure. The wafers are later sliced out of the ingot and then proceed to the ion implantation, masking, etching, oxidizing and metalization steps.

              If they are only yielding 1 sensor per 200mm wafer they cost per wafer is way more than $500.

            • Richard


              Indications are that the average yield is greater than that, but that some production runs may be as low as that. The author believed average yield to be 2.6 IIRC and that number may be higher today. Unfortunately, no one in a position to speak knows the actual numbers. The author of the linked article based his estimates upon his experience in the semiconductor industry. Remember that the article as written in 2006. Thom Hogans comments are base upon more recent information from whatever sources/contacts he has.

              It would be very interesting to see what Intel could accomplish. They at generally believed to have process control as good as, if not better than, anyone else.

              Until the yields come up there seems little likelihood of FX sensor costs dropping appreciably.

          • @Richard, great article about silicon yield rates. I will add this to your discussion, because I think you touched on it, but possibly misquoted the source. You mentioned stitching…and so I begin. There was article I read about Nikon’s sensor production some time ago when there was debates about where Nikon sources its sensor and wether or not they make their own. The article talked about how Nikon not only produces cameras, but how they also produce lithography and stepper equipment used in the production of the very sensors that go in their cameras. As a silicon fabrication plant is extremely expensive to own an operate, Nikon outsources their sensor production in a Sony FAB which is using Nikon’s lithography and stepper equipment in some kind of deal the two companies have. According to this article I read in a Silicon Journal (don’t remember the source now), Nikon’s lithography equipment couldn’t expose the very large 24 x 36 mm area in one shot, thus they have to 1/2 expose the silicon and then stepper the silicon wafer to the next portion and then expose the second half. (Hence the reference to Stitching) The difficutlies lie in precisely lining up the two exposures and the decreased yield rates which result. Canon on the other hand, posses their own FAB, and lithography equipment which can expose full frame sensors in one shot. Thus Canon has had the production advantage for quite some time. The article came out about the time of the D3X era, so it is quite possible that either Sony has upgraded their FAB or Nikon is sourcing their sensors at another location nowadays to reduce manufacturing costs. It would be interesting to know who has the FABs and what are the largest dies they can expose with their lithography equipment. I will also say this about FABs, they require lots of clean water in their processes, so a country which can provide that is probably a likely location for production. A friend of mine worked at Intel as a process engineer in the fabrication of Intel processors, at the newer FAB in Phoenix, AZ. He told me that Intel had difficulties getting purified water in PHX, because of all the minerals in the soil; they actually have to truck water into Phoenix which adds to their cost!

            • Richard

              @Dr SCSI

              Thom Hogan has discussed the stitching issue on any number of occasions and I believe that he has it on good authority that Nikon’s FX sensors, as produced by Sony, are stitched together. I have never heard any reference to a failure rate of sensors as a result of the stitching process as it relates to sensor yields, but everyone seems to believe that there must be a failure rate involved, which, at least to me, is a reasonable proposition. I believe that Sony/Nikon, or anyone else for that matter, would experience greater yields if it was not necessary to stitch together multiple segments of a sensor. How much? I have no idea in terms of actual numbers, but it seems to me that the allowable tolerance has to be incredibly small. That is just my intuition, however.

              The article I linked, authored by E.J. Peiker , was simply illustrating the difference in yields based upon sensor size, for a given number of defects per wafer. That alone raises the cost of a full frame sensor. The diagram in the article was to depict this principle. I do not believe that he intended to imply that Canon were able to make a full frame sensor in a single pass at the time of his 2006 article.

              Indeed, Canon have made public statements in the past that the APS-H sensor was the largest sensor they could make without stitching.

              I believe Mr. Peiker stated in one of the forum discussions that if a fab were to use a stepper which utilized a large enough die to make a full frame sensor without stitching that its capacity would exceed the world wide demand for full frame sensors. (I will try to look it up later.) The implication was that such an arrangement was not economically feasible.

              With the merger of the two Canon 1D lines into the 1DX I have come to suspect that Canon may have invested in the capability to manufacture full frame sensors without stitching. There is absolutely no information I have found to support this idea. It just seems to me that they would not have abandoned the APS-H sensor unless they had the production capacity to replace those sales with full frame sensors (not to mention the volume of 5D MK II bodies). The announced price of the 1DX is attention getting, but not really out of line with the prior 1Ds MK III when one looks at the increasing cost of just about everything and the current exchange rates of the Yen and Dollar. (Everybody who believe the government reports that there is no inflation go stand on your heads in the corner.)

              I would be very interested in any information confirming Canon’s ability to shoot a full frame sensor in one pass. With Canon having their own fab I imagine that little information gets out that they do not want to be known.

              Intel’s recent “Knights Corner” 50 core monster chip makes me wonder just how large a die they have the capability of utilizing.

              If Sony’s new fab is a 300mm fab I would expect that there will be a whole host of improvements that come along with the larger size wafer. Perhaps we will see some changes in Nikon’s full frame sensor process along with it. It occurs to me that a stepper large enough to shoot more than one APS-C sensor at a time would be useful in reducing production time and also be large enough to handle a full frame sensor die when its turn on the machine came up. There are a lot of interesting possibilities, but there is very little information known about Sony’s facilities other than that the company has very aggressive plans to increase their photo sensor production.

              Thanks for your comments.

              P.S. I have been mulling over the remark you made that the article you remember about Canon being able to shoot a full frame sensor in a single pass was in the time frame of the D3x. That might help explain Canon’s ability to price the 5D MK II the way they have…and by continuing to price the 1D MK series the way they have always been priced, and accepted by the market, were able to make a greater profit to pay off their new equipment even quicker.

              Thanks again.

            • photdog

              Not being deep enough in this technology, I wonder which attributes and features are influenced by the sensor itself and which are determined by the processing units after the sensor. I still feel that it must be more than coincidence that Nikon came up with cameras short after Sony did featuring the same sensor size (DX or FX) with the same number of pixels. Obiously Nikon could get a better output but still there is a suspicious similarity between Sony and Nikon sensors. Hence it is not easy to believe that Nikon does have their own layout.
              The D3x has a very well reputation and so many Nikon fans hoped for this sensor one day would boil down into a D700 body. As far as I know, there are 2 major rumors around, why this never came to pass:
              1) Sony wouldn’t give the sensor away for a camera priced in the area of a D700 to protect the sales on their own products.
              2) The D3X sensor was so expensive that used for a D700 successor would never be able to compete with the 5DmkII (price wise of course). That of course is also hard to believe, since we know how Sony priced their FF 24MP cameras.

              In summary there seem to be an intense dependency of Nikon from Sony.

            • Richard

              @Dr SCSI

              I have poked around in some of the Canon forums and it seems that there is a general belief that Canon have taken a decision to abandon the APS-H/1.3 crop sensor…and boy are they upset about it! Although is is an “in between” size, the advantage in “reach” to the wildlife crowd is important as well as the loss of autofocus function if they are forced to use their 500mm f4 lenses with a 2x TC on the FX 1D X instead of a 1.4x TC on the 1.3 crop body.

              There is even the suggestion by Thom Hogan (yes, he was on a Canon Forum) that Canon was pumping up their FX volume with the 5D MK II to spread out their R&D costs as well as to spread out production costs. (New one pass equipment????)

              No one has come out and said that Canon are actually able to image their FX sensors in a single pass. There is a Canon White Paper on the 1D MK III sensor which again says that the APS-H/1.3 crop sensor is the largest that can be imaged in one pass with existing equipment.


              There is another White Paper on full size sensors.


              Page eleven has a discussion of the number of sensors per 8″ wafer and mentions 20 defects per wafer. The paper mentions 20 full frame sensors per wafer as compared to 46 for APS-H sensors and “about 200” for APS-C sensors. It also talks about the extreme care required to avoid introducing defects on the full frame sensors during the many steps in manufacture because of the possibility of winding up with a wafer with no good sensors on it. It also talks about three passes to image the full frame sensor. There is a reference to page 53 on page 11 of 30 in the white paper indicating that there are three masks required for each stage of the full frame production rather than two people have been assuming. Note, however, that this was posted in August 2006 on Rob Galbraith’s forums so it may well be out-of-date.

              Stitching losses are still something which there appear to be differing opinions as to whether it remains a significant contributor to lost sensors. There was one comment about losing 3 out of 4 sensors due to this problem a few years ago and another indicating that the precision of the current steppers had reduced the problem for the most part. Who knows?

              There does seem to be general agreement that reducing defects is an area where important improvements in yields could be obtained.

              In short there still appears to be a lack of verifiable information about whether Canon have changed their manufacturing equipment to allow the manufacture of full frame sensors without stitching together multiple segments.

              It would be interesting to have five or ten minutes with the right people when they had to speak the truth.


            • @Richard, I found the article I read several months ago (maybe even years now) about Nikon, Canon, and steppers and lithography equipment. This was written by a company which closely examines silicon to help investment analysts and companies make decisions about sourcing silicon and production technology. Basically they indicated on Nikon FF sensors they could see the stitching that took place, whereas by Canon FF sensors, there was no indicitive artifacts visible which would indicate stitching was taking place. Direct quote from, “Canon appears to be the only lithography equipment manufacturer to offer an i-line stepper configured for 200/300mm wafers with a 50mm x 50mm field. It is only capable of ~0.5′m design rules, but that is consistent with the geometries seen on the Canon CIS we are analyzing. We have not seen any evidence of stitching in the Canon device, thus if they are using this stepper in their CIS manufacturing line, they likely have some cost advantages over the D3 foundry.”
              I’m enjoying the discussion by the way….I hope this article helps you see what all the photo bloggers aren’t telling you.

            • @Richard, as they say proof is in the puding.
              Here is a link to a Canon produced stepper capable of making their own sensors in just one pass!
              The largest field size is 52x34mm, and based on my other post with link to a article on Chipworks, it seems Canon has had an advantage for some time!


            • Richard

              @Dr SCSI

              Yes! Those are the same articles I found last night when doing some searching on the web. I think you saw my comment along with the links because of your comment about my observation that it appears Canon have had a substantial production advantage over Nikon for some time now. The picture of the Nikon FX sensor with the stitching artifacts was eye opening. I can not help but be suspicious that those artifacts do no good. It also seems clear that Nikon/Sony are making more than two passes to image the entire sensor which translates into reduce wafers per hour output which increases cost and on and on.

              When I saw the Canon Europe page where the specs were laid out in all their glory, I was simply awestruck! Canon makes that machine available to anyone who wants to buy it and Nikon struggles along with yesterday’s technology. (Did you notice the production rates of their machine?)

              The two Canon White Papers I linked put a lot of things in perspective even though they are somewhat dated. It appears that the Canon wildlife photo community has been reading the tea leaves with great accuracy for several years now that their much beloved APS-H sensors were going to go away.

              I am going to be very curious to see what Rob Galbraith has to say about the Canon 1D X when he gets his hands on one. As you will recall, he savaged the 1D MK III’s focus problems and was only a little bit kinder to the 1D MK IV in his review of the continuing autofocus problems Canon have experienced. If the new sensor performs as well as Canon asserts and they fix the autofocus Nikon will have a formidable competitor to the D4 when it is introduced. Independent testing will tell the tale, however.

              The Canon White Paper used a hypothetical 20 defects per wafer (which includes defects in the wafer along with defects introduced in the production process), but there were some comments in various discussion threads which seemed to indicate that something in the neighborhood of 10 or less was probably more accurate at the present time. If so, that would mean an average yield of 50% (or more) on a 200 mm wafer. No one has directly addressed yields on a 300 mm wafer, but I have to suspect that the yields would be at least as high as on a 200 mm wafer, probably higher.

              I realize that it is unlikely in Japanese culture that a company such as Nikon would swallow such a bitter pill as to directly purchase their rival’s equipment on which to manufacture sensors, but it seems obvious to me that Nikon needs to either have comparable equipment of their own (which we have not heard about as yet) coming on line at the new Sony fab being built or go in search of a fab somewhere which has Canon’s equipment and concoct a scenario where they put the job out for bids and fab X came in with the most competitive proposal for cost and ability to ramp up production (without mentioning what its production equipment may be) and so on and so on so as to save a little face. Even in Japan one would think that it would be better to suffer a bruised ego than to potentially lose the market. If Canon are able to bring quality full frame sensor bodies to market in volume at prices only hundreds of dollars above the price of DX based cameras Nikon could have a very bad time of it in all but the top end of the pro market.

              If their costs are as much lower than Nikon’s as I am suspecting, Canon has been reaping extra-normal profits even with the loss of sales due to the autofocus debacle. If they chose to do so, Canon could initiate a price war with Nikon at the top and put a great “squeeze” on them. They probably would not do so at the moment simply because it would be impolite to do so until after everyone has the problems created by the natural disasters resolved. I think that Canon’s long run average cost curve for full frame sensor production is bound to be much lower and flatter than Nikon’s. Simply put, that means that they can increase the volume of production while maintaining cost controls that a competitor with a steeper and higher long run average cost curve can not match. If this is the case, Nikon should be very afraid.

              It has been a most enjoyable discussion. I thank you for your comments and observations.


            • @Richard – “I realize that it is unlikely in Japanese culture that a company such as Nikon would swallow such a bitter pill as to directly purchase their rival’s equipment on which to manufacture sensors” — +1000
              Nikon won’t ever swallow that bitter pill, it would give a black eye to two divisions simultaneously! Just imagine, for publicity damage control, Nikon already had to make statments about using Sony FABs to produce their digital sensors. They re-ittereate publicly that it is their sensor design which is being produced on a Sony FAB in order to reduce their operating costs. And, they share their sensor technology with Sony in a symbiotic relationship; they both benefit from the marriage. The embarassment that Nikon suffered from someone identifing Sony as their source for sensors probably irks Nikon to this day. Unfortunately, nobody really said, “Hey, Nikon has a deal where Nikon provides Sony the Lithography Equipment and Steppers for their Fabrication plant, in exchange for production time of Nikon sensors in the Sony FAB.” That is closer to reality, but Japanese don’t toot their own horn either! Thus, I think Nikon will never risk the embarrassment of using a Canon FAB. Just think, the rumor getting out would say something like, “Canon must be REALLY good, because Nikon needs their expertise in Lithography and Stepper equipment, as well as in producing sensors for their cameras; why are you paying more for Nikon, just buy Canon and cut out the middle man!”
              I had previously read an article about how Nikon’s imaging department (lithography and steppers) was losing a ton of money, and it was the camera and optics division that was carrying the company. There was a strong competitor that they ignored, and they lost major market shares in the silicon chip imaging sector. Other than digital sensors, I can’t think of a need for such large exposure fields for intigrated circuits. But I can imagine internally at Nikon, the camera and optics division employees were probably thinking, “I wish the damned imaging division would get their shit together and provide us the equipment we need to make big sensors, so we can save face!” I imagine right now, Nikon is taking a loss on their pro body cameras. But at the same time, it is those same cameras that are in the hands of the world’s astronauts! It is the same pro body cameras you see at the Olympics, and on television in the hands of your CNN journalists who are reporting from the world’s hot spots! It is the same pro body cameras that Canon shooters are still lusting after! Although they probably take a loss on the pro cameras, the advertising costs are amortimized by the psychological and sublimenal messages which are planted in the heads of Millions of people who are buying a name brand. I guess you can liken it to the word “Volvo”. I am sure you immediately thought of “SAFE” in the same instance. I say Nikon and Canon to a non-photographer and they immediately think “Top Quality Cameras and Optics” and “Great Printers” respectively! I am sure Canon is also taking a loss on their top end cameras for the same reason Nikon does, appearances! Maybe one day Nikon will open their own fab, using their own equipment, to produce their own silicon. They could subsidise the cost by subletting the excess FAB capacity to other companies. Unfortunately, I don’t think Nikon has the excess of $1 Billion in capital to make such an investment just to make sensors for cameras which sell at a loss! I imagine Nikon will continue to make the very best pro cameras the industry will ever see, and leverage that exposure to help out sell Canon, Sony, Panosonic, Fuji, in the consumer camera market! I say, “Let’s get ready to rumble…..bring on the D4!”

            • Richard

              @Dr SCSI

              Yes, it will be a rumble leading up to the London Olympics.

              Speaking of Olympics, I recently heard a presentation by a photographer who went to the Bejing Olympics.

              He had been a Canon man, but the focus problems on the 1D MK III caused him to jump ship at the last possible moment…he read the manual for his new D3 during the 14 hour flight over there.

              “I got a Nikon camera
              I love to take a photograph
              So Mama, don’t take my Kodachrome away”

              (Paul Simon)

              Well, I’ve still got a Nikon camera!


      • “But on launch, expect to pay slightly higher than MSRP for first dibs, and MSRP for at least 6 months after, and maybe a 10-15% drop a year from then.” I wish 🙁

        Nikon’s equipment holds value really well, and it takes several years for the price to drop appreciably. And its prices don’t always go down. Just go on eBay or Amazon for Nikon SB800 or SB600. Both those flashes sell for more discontinued than they used to sell for new.

        Nikon makes great gear- but it costs.

        • @Robert Ash, the Nikon SB800 is an abnormalty. PROS have been paying more for them used because they believe they are better than the SB900 which replaces it. I own both and prefer the SB900; maybe I should sell my SB-800 for the price of a new SB-910! LOL
          As for the SB-600, I don’t know why anyone would pay more for that model. You are right though about Nikon holding its value well, but it sure does cost! In fact, most of my equipment has gone up in value since I bought it, because of the weak @$$ dollar. But one thing I have noticed, and it has held true over the last four years on B&H’s website; on launch date, the prices are high, but they typcially drop 10% within six months. Now with the disasters that Nikon has recently encountered, all bets are off because the supply chains have been significantly hindered. Low supply + High Demand = High Prices. That’s why we have seen the pro gear at such high prices lately. In anticipation of the D4, I sold my D3 just before the Tsunami hit Japan; I paid $3500 for it originally (refurbished) and sold it for $3500 two years later. If I would have sold it after the Tsunami, I probably could have gotten $4000 for it! Oh well… Anyway, there I am, no FX body but lots of FX glass. So I ran out and bought a used D700 for $1700 with only 25K exposures, to hold me over. I could probably sell that camera today for $2K, but I will wait for the D4 first!
          The wait for the new D4 has been agonizing for me, I miss the hell out of the D3.

      • Richard

        Here is an article with an interesting picture showing artifacts of stitching on a Nikon FX sensor. More importantly, the article references a Canon product with a 50 mm x 50 mm field size!

        And here is the product Canon offers for sale!

        It is listed as being capable of a throughput of 120 wafers per hour 300 mm) and 160 wafers per hour (200 mm).

        I guess this settles the matter of whether Canon have the capability to image a full frame sensor in one pass. The picture of the Nikon FX sensor shows stitching artifacts which settles that as well. It would appear that Nikon is at a very substantial manufacturing disadvantage.

  • FX DX

    I hope the body is not any smaller than D700. My FX zoom lenses and even 50mm f/1.4 are too big on my D90.

    • +1 @ FX DX! Anything less in size than a D700 would not make a good balance for FX glass.

  • JonMcG

    Ive been waiting for well over a year but I sincerely hope you are right and have to assume Nikon will not cut out this segment of their business.

    The 1DX, though I realize is priced in a completely different league sure looks like it could be Canon coming around to getting it right with the best balance resolution, performance, and noise… I’m a little worried the tide will be turning into Canon’s favor…

  • lasssal

    I gave a lot of thought to the posibility of D800 being 36Mp, yet retaining/beating D700 ISO capabilities.

    Then it accrued to me… The camera will switch to 18Mp, sort of a dual exposure mode from let say 1600ISO. This way, for landscape etc under good light we will have 36Mp and all the perks that comes with it. And for low light, 16 Mp with better ISO then D700.

    This is the only way Nikon can justify 36Mp camera.


    • bratvlad

      but than they would lose money by not making two camera’s instead.

      • The Manatee

        Disagree. This camera would be so versatile it could potentially sell more than two separate cameras combined.

        • Kerry33

          So versatile? Camera manufactures objectives are: make profit first,
          then 2nd-try to please their target market. Its not the other way round.

          Why the heck they wanna create juat one camera for everything?
          No way! At the end of the day, its about dollar and cents for them bro!

    • And how will they do this?

      Cropping? 18mp would not match the standard DX crop mode. DX Crop mode will instead be somewhere between 14mp-16mp

      Pixel binning? How? In order to pixel bin, you need to combine data from minimum 4 pixels — 2 horizontal and 2 vertical. Otherwise, the image will be distorted. So that leaves us with 9mp, not 18mp.

  • What is “substitute production”?

    • Jabs

      Probably a substitute Production location to replace the one now under water.

  • I think if D800 was going to be announced next week NR would have known by now.
    Also Nikon would have make more hoop la out of it. After all it would be a big step.

  • Can’t wait for full production to be back to normal. Have big job coming and I need the D800!!!

    • Keith

      “Have big job coming and I need the D800!!!”

      How can that possibly be true?

      And what kind of an idiot agrees to a job that depends on a camera that doesn’t even definitely exist?

  • JonMcG

    I’m actually not entirely sure that’s true, the sales market for a product like the D800 may not actually be significant enought to justify a rollout that big. Marketshare for $4K camera bodies with no lenses is only so big.

    I tend to agree if it were coming on Wednesday, I think we’d have more info but I don’t think Canon had that much fanfare for their announced flagship…

  • Gary

    OK let’s assume that the D800 rumors ARE true, and it’s 36MP.

    So (genuine question) if on day-1 they announce two version (with/without an AA filter) then:

    1) what uses of the camera would work best with an AA filter?
    2) what uses would work best without an AA filter?
    3) what other features might they put on the camera to compensate for the presence/absence of the AA filter?


    • NoFunBen

      that is the question people should be asking.
      I think i would be happy with the AA filter, i dont think i need more resolution then a normal 36. Landscape shooter want as much as they can get so for them the non AA filter may be useful.

      but once i have 36 i may find new things i can do that will use even more resolution.

  • Great question Gary. I’ll defer to others with more expertise.
    I really think a D800 will come out because below a D3s, there is a 3 year old camera (a great D700). If they decide not to upgrade this D700, there will be a HUGE gap in the FX line. D3s down to D7000? No, Nikon will unlikely let this happen and for 2 reason:
    1. Satisfy the prosumer (like many NR readers) and pros who want a FX camera
    2. Sell more camera and help the bottom line.

    Time will answer these issues. As for the timing, I leave it to NR to announce it.

    • bratvlad

      On the other side maybe that is exactly what they want to do, if you are a pro and need FF body got to pay big bucks. (and pros usually can afford that I think) And if you’re not, than there is not going to be a cheap new FF body. I might be wrong though

      • I think you also have a valid point bratvlad

      • Robin

        You ARE wrong.

        Its a multi polar market, and poor economy. No company is in any position to dictate the consumers, pro or otherwise.

        Over the few months, Nikon will release a cheap FF and a high end FF.

  • im’crapyinventor’too

    If you remember Nikon apply patent to replaceable sensor. I suspect the top secret will be: the 36 MP is optional since the sensor is replaceable.
    got it?…….Nikon still can insert the cheaper 16 or 18 MP and you just buy the optional 36 MP sensor if you like or if you can afford. I hope Nikon will do that, that will be a smart move…..

    • T.I.M

      To manage a 36mp sensor you need faster processor and more memory than 18mp sensor.
      So it’s not just about switching the sensor itself

      • im’crapyinventor’too

        it make sense if the camera has the capacity of 36 MP sensor as we see d800 rumors around….to those worrying about the huge 36 MP, Nikon can insert 16 or 18MP to a 36 MP capacity camera…..that goes with D4 too….that will be a smart move for Nikon…..i should call this the 36 MP the main line of Nikon sensor and the 16 or 18 MP will be the second line…;)….

        • NoFunBen

          the cost is not what is on the sensor (number of MP) but the size of it 36mm x 24mm.

        • Kerry33

          If that’s really happen, (Which I doubt it will) then only pros
          or rich people can afford the camera, cause its gonna cost us a bomb.
          Remember, nikon cameras already more expensive compared to equivalent models competitors such as canon, sony, pentax etc, by “only” edging in terms of build quality, & small extra featueres.

          And by enabling to switch from this to that and vice versa..(let alone switcting sensors!) Its very unlogical if there will be no huge increase in price..

          • Richard


            I realize that NR Admin strongly believes that there will not be an 18 MP D800, but there is precident. The D2H and D2X existed side by side in the same body, though not as user replaceable sensor bodies. If there is an 18 MP D800 I do not believe there will be a user interchangeable sensor or even a crop setting on the 36 MP body. It seems logical to have a modular body design that is capable of accepting different sensors and/or image processing units at the time of manufacture. Unless there is a reason to change the body design, it can shorten the product development cycle as well as reducing costs to leave well enough alone on as much as possible on the chassis.

            The question in my mind though is whether Nikon believe that there is a need for such a camera at this position in the product lineup. It seems more probable to me that Nikon believe that the (presumed) D4 will better fill the needs of the sports/action/photojournalists.

            At some point Nikon will have to provide an alternative to Canon’s 5D Mark II/III lineup or concede that market segment altogether. I am not terribly optimistic about a Nikon product for that segment unless the rumored price of the D800 is completely wrong. After all is said and done Canon are on the verge of introducing the third generation of the 5D and here we are wondering what Nikon are going to do.

            Perhaps all will be revealed soon. I hope the news is good.

            • There could be another 18MP full frame camera later on but not now. If the D4 is 18MP, Nikon may use this sensor in the future.

  • Hi, I have a question for the sensor design experts here.

    I’m wondering why Fuji was considered as “cheating” by many when they cited the Foveon in the past as being 12 MP when it had 4 MP with the R,G and B stacked on each other. My understanding from reading posts here awhile back is that regular sensors have R,G and B laid out side by side. Why is laying it out that way considered ok but stacking the sensors is not ok?

    Also, if I’m misunderstanding sensor architecture please correct any misstatements I’ve made in asking this question. Thanks!

    • The context for this question, of course, is why would it be any worse to stack 36 MP into 12MP worth of pixel sites (maybe larger ones, which could be better for IQ) rather than 36 MP worth of smaller ones?

      Also, I believe the Foveon counted 2 pixels as one, not three, and it also had a large pixel / smaller pixel pair. So it counted its 6 MP as 12 MP. Sorry, it’s been a long time since that camera was around 🙂 So it makes sense why people didn’t go for that.

      But what about stacking RGB on each other? is that physically feasible even?

      • NoFunBen

        Foveon counted 1 pixel as 3. also 1 pixel could see all 3 colors of light. but Foveon was bad in low light. we dont want to lose our low light performance.

        Foveon looked at the light level in 4 million places. it also looked at color for r g b in each of the 4 millon places, they called this 12 MP.

        a regular (Bayer) sensor looks at light at all 12 million places, but only one color at each place. red and green on one line then blue and green on the next.

        our eyes see resolution mostly in the green color so green is looked at twice as often (a green sensor site for each red one and a green sensor site for blue one ) so a bayer sensor of 12 MP has 6 million green and so it beats the Foveon of only 4 million green sensors.

        but a 8 MP bayer sensor has 4 million green sensor sites same as the 4 MP Foveon, this is where the 2 to 1 you are hearing comes from.

        in the end you still have light being recorded at only 4 million places and you have poor low light noise levels.

        we dont want any of these things. maybe some day Foveon will be a competitor but that day has not yet come.

    • Because the Fuji sensor didn’t result in a 12MP linear resolution image, it was still a 4MP image. Same thing with Sigma’s “45MP” Foveon sensor in their SD-1 – It’s only really 15MP in actual image resolution.

      • One of the biggest advantages of the Foveon design was the fact that you could delete the Anti-Aliasing filter. Now that Bayer sensors are approaching very high resolution, you can do away with the AA filter on Them too, because moire is less of an issue. Hence the rumored deletion of the AA filter on the 36mp D800.

      • “Because the Fuji sensor didn’t result in a 12MP linear resolution image, it was still a 4MP image. Same thing with Sigma’s “45MP” Foveon sensor in their SD-1 – It’s only really 15MP in actual image resolution.”

        The Fuji sensor was 6 MP (see my corrected number above, I’d forgotten there were 2 pixels per slot, not 3).

        I think what happened is that Fuji and Sigma counted and count each lens as one pixel, whereas everyone else counts 3 lenses — R, G and B — as one pixel.

        If that’s not the case then someone please correct my statement.

        It’s been very useful “thinking out loud” here on the forum by the way 🙂 Many thanks to all the informative posters here.

        • John Richardson

          Yep. One of the best threads so far.

        • Yes, you’re correct. I stand corrected; It was a 6+6 sensor.

  • Mamu Line Clear

    OOOOhhhhh yeaaaaaahhhhh, where’s my D800?

  • The Manatee

    What has Digital Rev’s accuracy been in the past? Anyone know?

  • Stilld70

    Mabe it is a bit off topic, but I wonder why they do not produce a square sensor, so you do not have to turn your camera. The production cost should be +/- the same, the image circle of the lenses could cover the whole sensor and you have all kinds of freedom regarding your format 5/4-3/2 and so on.

    • @Stilld70 – I think your question concerning size is not far off at all. I was just thinking yesterday while reading this forum about sensor size and I wonder if at any time Nikon has considered designing a medium format digital camera. I think it would be great! I also wouldn’t mind shooting with a square sensor and agree with you totally regarding the freedom to format in 5/4 or 3/2 or whatever you really wanted at that point. Thanks for bringing that up, turns out I’m not a lone soldier on that subject after all.

      In regards to the D800, I’m excited and look forward to whatever Nikon releases. The 36MP sensor will really be something if it’s true! I’m in a position to upgrade to a new camera so we’ll see what happens. In the meantime I’ll be shopping for a D700 to hold me over (my d40x needs the shutter replaced).

  • alvix

    looking at all those super duper hi ISO numbers…like 102400..204800 ..whatever ..I think we should start using dB’s +15dB +21dB’s etc…

    • aonexia

      That’s actually a pretty sensible idea. Soon we’ll be into 7-digit numbers! You probably wouldn’t want to mark it as “dB” though as I think many people would struggle with the concept of Decibels being used to measure sensor sensitivity.

    • aonexia

      Oh Oh, then can we put some optional attenuators between the sensor and ADC?

      So I can leave my ND8s at home and have -20dB exposures? 😉

  • zack

    Does anyone know (based on rumors) HOW better video are we’re going to get in D800?
    Say, we compare it to D7000 which has very good video. Are they going to up the bit rate? Better codec? (unlikely)

    • Sankofa

      I hope the ne camera will provide at least 50 fps full HD but If the specs are right its 30fps.

    • Anon

      Dude, who cares the video and its specs? I’ll get one even there’s no vid, just for the sake of 36 mp.

  • I saw a post claiming a french magazine is predicting both a D800 and a D900. 18MP for the D800 for $2500 and 36MP with no AA for the D900, for $4000. I don’t know whether to believe this at all.

    • Chris Ni
    • Gary

      Seems logical, but wouldn’t a $4,000 price tag suggest this would be in D4 territory?

      If Nikon wanted to hike prices this much, surely folk would start to consider alternative brands? After all NAS only affects he obsessives. Nikon would have to DEMONSTRATE the benefits of such a price hike and not just HYPE the features.

      Fog example, an option for my car is a “a heated REAR seat” (i.e. feature) but as my wife and I seldom have rear seat passengers then there’s no benefit in us buying that option.

      So, IF this rumor is correct, then I’d want to understand the benefits I’d get from this $1,500 feature compared with, say, a new lens!

  • Mimmo

    Ok … can I have my Nikon D850?

  • This thread is EPIC WIN!

    Great knowledge and conversation can be accomplished when we’re not busy bashing one another; This thread is an example of what we can do when we’re all playing nice. Let’s keep these kinds of threads going, guys. We can accomplish so much more if we stay positive and help contribute to the community.

    • raplh

      Help the community? Contribute to the community by leaving comments about Nikon products? Really? Are we feeding the hungry? ha…

    • Jabs


      Yes, true and all we need now are Instructions to get files off crashed drives and OS systems gone awry.

    • +1 🙂

  • Mauro

    What happens with the D300s replacement?
    When is that coming out?

  • Mike

    What is going on? p a p a r a z z i n c . c o m

  • Well, as I’ve shared here before, if the D800 comes out with specs predicted and with great low-light performance (and video :)), it will be the digital camera I’ve been wanting for some years now and should last me well into the future. Would be fantastic!

  • Tell me more about d800 please


  • Back to top