Nikon's official statement on the D3X sensor

Discussion in 'Nikon' started by ellis_vener_photography, Dec 9, 2008.

  1. You can now stop speculating. While Sony is doing the fabrication the D3X CMOS sensor is completely designed and engineered by Nikon. To make it crystal clear: It is not the same sensor in the Sony A900. Here is the full text of Nikon's statement as released to me this morning:
    "The Nikon D3X’s 24.5-megapixel FX-format (35.9 x 24.0mm) CMOS sensor was developed expressly for the D3X in accordance with Nikon’s stringent engineering requirements and performance standards, with final production executed by Sony. Featuring refined low-noise characteristics, 12 and 14 bit output, Live View capability and more, the D3X’s unique sensor design was carefully blueprinted to perform in perfect concert with proprietary Nikon technologies including EXPEED Image Processing and the Scene Recognition System. Meticulous efforts allowed the sensor to become one of the many essential components and technologies which contribute to the D3X’s superior image fidelity."
     
  2. Could this not be paraphrased as "It's basically the same, but we we need to make it sound special for this press release"? I mean, would anyone really expect Nikon to say 'Yeah, you can get the chip that comes in our new $8000 camera in a Sony that costs way less"?
     
  3. It reminds me of the old 70-300 ED, when somebody pointed out that Tamron was selling an identical lens. :)
     
  4. Rob, you can believe what you want or make a mockery of the facts if that's the kind of thing you enjoy. If Nikon actually used the A900 sensor in the D3X, do you honestly believe that Sony would allow Nikon to give such a press release?
     
  5. This is useful, thanks for fishing this out.
    Although it makes it clear that the sensor has some aspects that are specific to the D3x only, it does not state that it's a Nikon design so it may still be a derivate version from the Sony sensor. They did something similar with D70, if you recall. If memory serves me right, the D70 had a combined mechanical/electronic shutter mode that was unique to Nikon.
    Note that "Nikon requirements and performance standards" does not mean much: it is common practice when a company subcontract the design or the development of a component to impose requirements and performance standards on the contractor.
     
  6. I never understood where that claim or argument came from. The Nikon & Tamron 70-300 lenses looked nothing alike and had different filter sizes. I vaguely remember seeing internal diagrams of the lenses and they were different as well.

    OK, back to the D3x discussion already in progress...
     
  7. Clay, the article seems to be talking about the D3 not the D3X.
     
  8. Eric, sorry to go off topic, but believe me, the now discontinued Nikon 70-300 4-5.6 AF ED (the lens that replaced the 75-300 AF) and the Tamron 70-300 AF ID are exactly the same design. I have both because I use the Nikon and I got the Tamron in gift (I'm not selling it since the resale value is miserable) and if you look at them side by side you will notice not only that they have the same 62 mm filter thread, the same dimensions and design but also the same internal movements, building construction and the like. At the time I recall somebody saying that the Nikon had a better quality optics because of ED glass and superior coating, but this is my memory. For this reason I am not surprised of the above statement.
     
  9. Thanks Mike,
    How about we split the difference ?
    Nikon also makes machines called 'steppers' that actually manufacture
    the sensors. Nikon's biggest customer for their steppers is Sony, ha !
    So, who made the sensor ? Who cares, long as it works.
    Best regards,
    /Clay
     
  10. "Who cares, long as it works."
    Exactly, as long as it does the job, it's perfectly matched to the camera and it's up to Nikon standards for a professional camera, who cares about the supplier. The manufacturer could well be ACME Inc., as long as it is not the crappy stuff they sell to Wiley E. Coyote. ;-)
     
  11. I don't see what is so abhorrent about this process. when I designed hardware, I evaluated existing components and integrating them partially into my design, versus the problematic costs and time-constraints of not doing so. a good design is extensible, and surely Sony in concert with Nikon (and others) created an easy migration-path for designers with specific architectural requirements. I recently designed a rotating platform to align the solar-panel arrays on my sailboat with the sun, as the sun moved and the boat moved relative to it. I had no problem using a micro-controller from Motorola that incorporated the hooks I need for my customization, instead of attempting to plod through my inventions.
    not the same I'll admit ... but I don't see that the issue at hand amounts to much.
    daniel taylor
     
  12. Interesting that the number of horizontal/vertical pixels and the sensor size (35.9 x 24.0 mm.) are exactly the same!
     
  13. I think the whole thing started when people who wanted the D3x, but couldn't afford it, began whining about it, saying that why could Sony sell the A900 for so much less and it has the same pixel count?
    I could care less, I want a Nikon, not a Sony. It's like saying that the Dodge sedan is a Mitsubishi, because it has a Mitsubishi engine in it. No, it's a Dodge, not a Mitsubishi. Just because the Sony A900 has a similar sensor doesn't make it a Nikon.
     
  14. Agree with Dave here. The BMW Mini is not a Peugeot, even if it shares the same engine. What matters in a camera is not the single component, it is the final result.
     
  15. ' it does not state that it's a Nikon design so it may still be a derivate version from the Sony sensor"

    It is not derived from a Sony design. Sony is doing the fabrication work only. But as Illka said: you'll believe what you want. Ignorance is bliss as the saying goes.
    Think of it this way: Sony has large fabrication facilities that need to be kept busy. From a financial point of view it is no problem to say this assembly line over here makes the chips for Sony cameras and this one makes the chips for Nikon cameras.The two design and engineering are kept seperate for various reasons mostly having to do with intellectual property right issues. For all we know the Sony plant may also be fabricating the RED camera sensors as well. We know that up until recently Canon was a camera sensor customer as well for point and shoot cameras. Canon is (or possibly was until very recently, depending on who you talk to) a a long time big customer of Nikon's for steppers as well. Companies in Japan ware set up very differently then they are in the USA. Yes they compete but not stupidly.
     
  16. "The Nikon D3X’s 24.5-megapixel FX-format (35.9 x 24.0mm) CMOS sensor was developed expressly for the D3X in accordance with Nikon’s stringent engineering requirements and performance standards, with final production executed by Sony."
    Yes, I have no doubt that the sensor was developed "expressly for the D3x in accordance with Nikon’s stringent engineering requirements and performance standards, with final production executed by Sony". It's not a case of the Sony A900 sensor being used in the D3x... it's a case of the D3x sensor being used in the A900! That is almost surely the case. It's a clever statement from Nikon that attempts to lead the public into thinking the sensors are different. Somehow I doubt it. No doubt the A900 was released prior to the D3x to judge public opinion, give the Canon 5D2 a scare and to test the sensor in action before applying it to a Nikon. (That's my theory anyway).
    Nevertheless, I still think it's a pointless debate really. There's more to the camera than the sensor. The firmware, image processing, design etc will have just as much impact on the end result as the sensor.
     
  17. "It's not a case of the Sony A900 sensor being used in the D3x... it's a case of the D3x sensor being used in the A900! That is almost surely the case. "
    Jamie, it's best not to make statements that are not based on facts you can back up with references.
     
  18. And I bet the Sony A900 is way below the D3X in terms of overall quality, sturdiness, speed and design. How many $$$ below is hard to quantify, but still.
     
  19. The A900 and the D3X are aimed at very different markets with different goals in mind. As I generall yunderstand it The ranking in global D-SLR camera sales is a virtual tiie between Canon and Nikon with Sony a distant 3rd. Sony wants to close that gap.
    For all any of us know the A900 is being retailed at or for far less than actual manufacturing (includding marketing) costs in an effort to gain market share. Once you have people buying yourcameras you also have them (for the most part) buying your lenses, speedlights, etc. Canon takes care of this n the high megapixel end by having two cameras with roughly similar pixel count (EOS 1Ds MArk 3 and 5D MArk 2), But beyond the mp count the to Canons are very different. Sort of like Nikon with the D3 and the D700. The small people who actually understand (and the larger market segment who thinks more $$$$ means more camera) what they are getting with an $8,000 DSLR, whether it is the D3X or the 1Ds Mark 3 will gravitate to those cameras.
     
  20. I guess the only answer is to buy one of each and take them apart and reverse engineer the sensors.
    Any takers?
     
  21. Why bother, Jim? In fact, why bother with the entire issue at all?
     
  22. It's really comic that people think the entire camera is the sensor. It's like saying an entire PC is the CPU. People are getting all bent out of shape over the fact that Nikon outsourced the manufacture of a sensor they engineered and designed in house to a facility operated by Sony. Folks, this happens all the time. Companies outsource production for everything. Boeing outsources the manufacture of most of their airplane parts to other companies (including entire wing, fuselage, and tail assemblies, which arrive at Boeing fully assembled on the top of a rail car or inside a large cargo plane) and merely assembles the final product in their own factory. Yet it is still called a Boeing airplane, despite the majority of it being produced outside of Boeing. Yet it is a Boeing airplane because Boeing engineered and designed it, and they set up the specs for which it was to be manufactured. Same with Nikon and the sensor for the D3x. Just because they have Sony manufacture it doesn't mean Sony has the right to put the same part in the A900. That would be grounds for a lawsuit.
     
  23. These 737 fuselages were photographed on a moving train passing Boeing Field near Seattle, where they will later fly to for flight testing and delivery. Photo taken 11-23-2008.
    [​IMG]
     
  24. "If Nikon actually used the A900 sensor in the D3X, do you honestly believe that Sony would allow Nikon to give such a press release?"
    Nikon is a billion dollar customer account for Sony Semiconductor.
    Olympus bought a few tens of thousands of sensors from Kodak. Because of this, when Olympus asked Kodak execs to stand up beside them at a press conference and say "we have been working together on the four birds thirds system", Kodak snapped to...
    If Nikon told Sony they wanted a dozen execs to swear up and down that Nikon and Sony together hand coated every sensor with green tea mochi, Sony would say "mmmm.... mochi."
     
  25. "...it's a case of the D3x sensor being used in the A900! That is almost surely the case."
    It's possible, but somehow, I doubt it. Given people's access to information, it'd be an unwise move for Nikon.
    Personally, I believe it the design and quality tolerances that matter, not who fabricated it. I would suspect the tolerances and specs for the Nikon sensor are higher than what is housed in the Sony, even if it is the same basic item.
     
  26. You say the following:
    "While Sony is doing the fabrication the D3X CMOS sensor is completely designed and engineered by Nikon."
    But the quotation you give says:
    "The Nikon D3X’s 24.5-megapixel FX-format (35.9 x 24.0mm) CMOS sensor was developed expressly for the D3X in accordance with Nikon’s stringent engineering requirements and performance standards, with final production executed by Sony. "
    That does not say what you said, so what is your source for the D3x sensor being "completely designed and engineered by Nikon"? Like many cars, the A900 and D3x sensors could well be the same basic pattern, but the D3x version has been modified to suit Nikon requirements hence they can say "developed expressly for the D3X in accordance with ...". and they are not lying.
    But this arguing is no more than weeing in the wind.
    The only way to know how good the D3x sensor is will be through testing. B. Rorslett will have one and hopefully will report his findings on the Nikon Gear web site. I think he can be trusted to give a reliable and honest account.
     
  27. because I have had other talks with Nikon about it at the time of the D3/D300 (directly with the engineers inolved, in Tokyo and Sendai) launch and with Nikon USA representatives prior to the official announcement of the D3X.
    I am as honest as Bjorn is (maybe not as techy). You can cross verify all of these statements with Imaging-Resource.com
     
  28. I find the (potential?) commercial confidence issue interesting. Say Nikon makes some great advance in sensor design, and then gets Sony to make the sensor. Is this not running the risk of letting your competitor have a first hand look at how they could also make their sensors better? How do these things work in an arrangement like this?
     
  29. Ellis: I was not for one minute questioning your integrity, merely wondering if you based your conclusion on the quoted text alone. Anyway, thanks for the information.
    So are you saying that Sony had no input? And do you have any explanation for why the Sony A900 and Nikon D3x sensors share so many similarities such as pixel count and other details? I think there are architectural similarities. Not that it is that important ... :)
     
  30. Leif,
    No I am no saying that. I am certain that Nikon worked with Sony engineers to see what was possible with Sony's CMOS making capabilites and then took those baseline qualities in a proprietary direction.
    "Is this not running the risk of letting your competitor have a first hand look at how they could also make their sensors better? How do these things work in an arrangement like this?"
    Non Disclosure agreements that go both ways. Trust. Long term personal relationships. Successful capitalism sometimes requires collaboration even with erstwhile competitors.
     
  31. Yeah trust.
    Anyone remember when Kodak was making instant film packs for Polaroid? They got into a ton of legal trouble when they stole the Polaroid technology and used it in the Kodak instant cameras. After a big lawsuit, Kodak had to pull all their cameras out of retail distribution.
    Trust.
    &nbsp
     
  32. This is really a discussion about the relative prices of the Sony A900 and Nikon D3X. Why should the Nikon cost so much more than the Sony when it uses the same sensor? If you recall all film cameras can use the same film but were a wide range of prices. The issue I think is the level of price difference. I just checked and in 1986 the top of the range Canon F1N was $810 vs $350 for the cheapest - the AE1P. This is a 2.3x ratio, whereas the Nikon to Sony ratio is perhaps 6x. While the construction and physical capabilities of the Nikon will be better than the Sony the difference is hard to explain. I first noticed this Digital premium when Canon replaced the EOS1V with the 1D series. The 1V was under $2000 and the 1D was 2 to 4 times as much. Apart from the display and sensor these cameras are almost identical.
     
  33. Spearhead

    Spearhead Moderator Staff Member

    The statement is very obvious marketing double-speak. They are being very careful not to say that they designed it. There's absolutely no reason to phrase it this way if they did the design.
     
  34. This thread reminds me of why I rarely come here anymore. People arguing about things they really don't know anything about, which wouldn't make a damn bit of difference in any case.
     
  35. I agree with Mark Ci
     
  36. It doesn't matter - if Sony made the sensor and we get great images, that's fantastic! But really... if it's a Nikon design and Sony's manufacture, then they an easily say it just like I did. Which they didn't. But once again: whoever designed it doesn't matter.
    BTW regarding the trust issue, we have to remember that the Japanese do things differently than in other countries. Ellis put it well.
     
  37. The most interesting part of this discussion for me is that until now Canon has had far greater commercial grunt than Nikon - with all it's other divisions - printers, photocopiers etc. Nikon, on the other hand has been far smaller (although still large) without the same R&D budgets etc.
    If Nikon needed a large scale manufacturer to help with their products (whether in manufacture or design) then why not get into bed with Canon? - Surely that would put an end to the "Canon vs Nikon" argument once and for all.
    Obviously they were never going to do that, and so I find it strange that they would choose to get into bed with another DSLR manufacturer who is in direct competition with them. Sony is of course, a giant of home electronics - larger even than Canon.
    Of course with the credit crunch, Sony are feeling the pinch and as stated earlier, need to keep their product lines busy - but then again, I'm guessing Canon's in the same boat!
    So are we seeing the beginnings of a partnership between Nikon and Sony, with Nikon engineering Sony's DSLR lineup (and Sony manyfacturing Nikon's lineup) Something along the same lines as Jaguar and Ford? If so, then the result could swing one of two ways.
    I personally feel that Jaguars have been "cheapened" by Ford manufacture, and have a gut feeling that the Nikon line might go the same way. Of course, the result would be a formidable competitor for Canon, and the cost reductions resulting from Sonys intervention would surely bring down the price of DSLR's further, in the long run.
    Of course, it could go the Lamorghini / Fiat way. (Fiats are ten times the car they were before that hookup and Lambo's don't seem to have suffered at all).
     
  38. Guy, Sony has been manufacturing sensors for Nikon DSLRs since the beginning of time. It's not a "new" partnership.
     
  39. This is way off topic but Lamborghini made tractors before he made sports cars. He sold his tractor business to Fiat (and if that technology transfer improved Fiat cars tenfold, what can I say?). The sports car business also changed hands several times and is now owned by Audi.
    I don't think it says anything about the Nikon/Sony deal, but maybe something about how well informed this discussion is.
     
  40. To me, this is a careful smart-lawyer rephrasing to go around the taboo word "Sony"
     
  41. I wonder if Sony+Nikon will eat Canon DSLR sales in years. I wonder if Nikon will be eated, too... if so, please keep the Nikon F mount... ;)
     
  42. Lamborghini is not Fiat. It belongs to Volkswagen. Fiat owns Ferrari and Maserati.
     
  43. Luca,
    Now which one is D3X, fiat, volkswagon, Ferrari or Lamborghini?
     
  44. I told you. The BMW Mini and some subcompact Peugeots share the same engines, which are the result of a joint development project. The last generation of the Mini here was offered with a common rail Diesel engine manufactured by Toyota and the same 1.6 gasoline engine was used by Chrysler on the European PT-Cruiser. Problems? A car is not only the engine.
    Same applies for photo equipment. Developing a digital sensor and setting up the production lines for it requires a lot of money. Then if you are lucky, you recover the investment running the production lines at full speed for the few years that this piece of equipment will be used, before introducing the next one. I see nothing wrong in using the same sensor for different cameras of different brands, as long as each brand builds its own camera around the sensor. As I saw no issue in Nikon using a Tamron design (or viceversa) some years ago. At the end you have different products targeted at different markets. Most of the issue here for me is coming from two things. The first is that the Nikon is outrageously overpriced, even for the pro or the rich amateur. The second is that for many years marketing departments have pushed the concept that a digital camera IS the sensor and the megapixels. So, at the end of the story I think cameras are intended to take pictures and the answer will come the firs time somebody will do a side by side comparison of both cameras, and answer the fundamental question: do you see a difference in the pictures and is this difference worth the difference in price. In other words, assuming that the Sony and the Nikon have the same identical sensor, do you get something more buying the Nikon that is worth the price?
     
  45. The first is that the Nikon is outrageously overpriced,
    Sound like you're talking about a user with 2-3 lenses. Whether the D3X is overpriced or not, depends on the image quality it can deliver, and the system its tied to. The actual street price is unknown and the list price is the same as Canon's current 1Ds Mk III. Sony's accessory and lens lines are much more limited and I don't have any information about their support network (apart from the fact that the same company's laptops have essentially no support and IMO their best use is by having them recycled). Sony's inability to even begin to approach Nikon and Canon's market share is simply a testament to the fact that just by being a sensor manufacturer doesn't carry the product line by itself if they don't know what to put in front of the sensor, and what to do with the data from the sensor.
    In other words, assuming that the Sony and the Nikon have the same identical sensor
    It was just said that the D3X sensor is exclusive to that camera. Why is it so hard to take it in? In any case the optics in front of the sensor and the image processing seem to have a considerable effect on the image quality, so the possible similarity of the sensors without the optics and processing (which could be very superficial) doesn't mean that image quality will be similar - if past experiences are anything to guide us, there is likely to be a world of difference. But anyway, in a couple of months we should have all the data that is needed to see why the Nikon is priced where it is, or if it's not correctly priced, the price will go up or down.
     
  46. My word is "assuming". Even assuming that the sensor is the same, considering that a camera is not just a sensor, is the price difference worth or not?
     
  47. "In other words, assuming that the Sony and the Nikon have the same identical sensor"
    The A700 and D300 have 'the same sensor' and yet the D300 performs much better at high ISO. The truth is that t he sensor is not the same, and Nikon added Nikon specific features to the sensor and the DSP pipeline . G iven information from Ellis, the D3x and A900 sensors are related, but we will probably never know quite who did what.
    There are a lot of people dissing the D3x without having used it. We need to w ait for thorough tests. Pricewise, the camera is priced according to what Nikon thinks users are prepared to pay. If no -one bites, the price will fall.
     
  48. Surely there is some common technology in both sensors, but as many people said, who cares? Are you going buy you own sensor and pimp you D700's to have 24 mpix? I think not, so what matters is the final package. Now it can be discussed whether the D3X is overpriced or not, but it's only relevant for someone who actually plans to buy the thing right now. Are you going to wait 12-24 months anyway? Then what the price is now is not very important.
     
  49. Dave is right - a camera must be viewed in its totality. Much work in fact comes post-sensory and is software (having a better noise reduction method than others, having a better tone mapping algorithm than others, having a better de-mosaicking method than others). So, they may have the same sensor, but results may be entirely different because of different image processing methods. PS: of course, also different hardware components after the sensor, e.g. with lower noise, are important, too (depending on the implementation of the aforementioned image processing methods).
     
  50. As for being "overpriced:" In relation to what? Another camera of the same MP? Well, if it's just about megapixels - I guess all DSLR's are overpriced. I can just go buy a 12mp P&S, and it will do the same thing as my D300, right? It's the same resolution - what else matters?? How many of us spent the extra on Velvia rather than just buying the cheap film? How many are out here proclaiming the injustice of companies like Mamiya and Hasselblad. And, who makes their sensors?
     
  51. WHO CARES.... Nikon users have two choices for a 25mp camera. Buy the D3x, or buy a Sony a900 (or whatever) and a bunch of new lenses, flash, accessories, etc. Either way, you're going to spend $8,000. Sorry to get snappy, but you're going to have to get over it.
     
  52. ShunCheung

    ShunCheung Moderator

    For decades, the most expesnive 35mm film cameras incluidng the Leica R series, Nikon F series, Canon's EOS 1 whatever ... use exactly the same sensor (film) and the point and shoots or even some disposible cameras. In fact, you can shoot half a roll in a point and shoot and then move the film to a Leica R9.
    How strange that all of sudden "using the same sensor" is the biggest injustice in the world.
     
  53. No 35mm film camera intended for normal use cost $8000. The difference between the F6 and the D3X is mostly in the recording technology (ok, F6 is a previous generation camera, but that's beside the point); a substantial part of the cost of the D3X is clearly in the digital part. Therefore, if there had been a chance that the A900 gave image quality similar to the D3X with the best lenses available for the cameras, respectively, then the $5000 premium could be put into question.T he samples that have been put online so far would suggest to me that the D3X is correctly priced; not aggressively like the D300 and D700, but nevertheless in line with the image quality, features, and the competition. Production models may be better still. Fortunately the bleeding edge in quality doesn't stay still for long and these technologies propagate to lower price categories. Which is the most important thing that people complaining about the pricing should remember. Whatever the D3X can do, will soon enough be commonplace.
     
  54. I agree with David Ahn.
    Dave Lee, if you go north a bit to Interbay neighborhood in Seattle, you can get right on top of those fuselages. I see them quite often in the Interbay rail yard. Really interesting the fuselages mixed in with other rail cars. Tom
     
  55. "the D3X’s unique sensor design was carefully blueprinted to perform in perfect concert with proprietary Nikon technologies including EXPEED Image Processing and the Scene Recognition System."
    LOL, I'm glad this new sensor works in perfect concert with their own EXPEED systems, as I was getting a little worried! phew....
     
  56. This is the official stmt ... :)
    http://www.photo.net/nikon-camera-forum/00Rj0k
     
  57. Who Pays list for equiptment thesedays?
    Canon 1Ds III is buyable tonight at B&H for $6499 w/ free delivery to your USA Door.
    I've ALWAYS thought these $8,000 dslrs have been the high profit items for Canon for years. Kinda like $50,000 pickup trucks. Sony may be making the A900 their loss leader but nikon's pixie dust premium and mount is what 6 ounces of gold or platinum will run you today.
    If having a Sony Made sensor in your unissued $8,000 dslr is a turn off then buy a $4,099 D3 or a $2379 D700 instead. Or buy one of each and a couple ounces of gold for the future. I've read these 12mp in D3 and D700 are Nikon designed & produced sensors and at almost double frame rate & 1/2 storage space: What A Bargain! I'll guess you'll need something like CS4 for D3x too, just like the 5D2 buyers are realizing now. Anybody priced CS4 upgrade?
     
  58. Anyone who thinks the D3X is overpriced is simply a whiner. That's the basis of this entire thread. The people who bring these things up are whiners. Face it, you'd love a D3X but you can't afford it. Get over it. Go buy a Sony A900 and be happy.
     
  59. The comparisons to film, and claims that all film cameras used the same "sensors" are specious nonsense. The people making these claims either weren't competent at shooting film, or have darned lousy memories.
    With a film camera, you could change films to suit your purpose. You wouldn't have used Tech Pan or Velvia to shoot an average wedding any more than you'd expect landscapes with highly saturated colors from Vericolor S. At least with color print film, you also had a fair amount of control via the exposure you used -- it doesn't take a great memory to realize that if you wanted saturated colors and all you handy was VPS, a +1 (or 1.5) exposure would help quite a bit. It still wasn't like Velvia, but at least it wasn't as washed out as VPS was at rated speed.

    With a digital camera (at least ones like this) the sensor is an integral part of the camera -- getting one sensor for landscapes, another for portraits and a third for night shooting would mean buying three (expensive) cameras. You can't play many games with exposure either -- a +1 exposure won't give more saturation, just blown highlights.
    As for the rest of the camera: when shooting in raw mode (would anybody be insane enough to buy a D3x to shoot JPEGs?) the processors, etc., just format and shuffle the data. If they affect the image quality at all, there's a problem (cf. the Sony Alpha 700 prior to V4 firmware). Likewise, for a serious photographer the focus point and exposure the camera chooses are really just hints, not final decisions -- if the photographer is at all competent, they won't affect image quality to a significant degree.
     
  60. Everyone seems to be missing the point. It's not the sensor that counts, it's the lense you fit in front of it.
    I've bought a Canon EOS 400D (Rebel to those of you in the USA) along with a Sigma 18-200 and find that on wide angle shots with a horizon, ie. the sea, I get a smiley horizon, now I'm sure this isn't the sensor, or the camera, but the lense.
    I'm a Nikon fan due to having a complete range (15mm to 1000mm) lenses which I bought for my still operational F2's and they still take exceedingly good pictures. I've just ordered a D700 with 24PC-E lense (not yet arrived, as Nikon assures me that I can still continue to fit my old lenses to it).
    Build quality is also a factor to take into consideration.
    Once I bought a Tamron 75-250 (back in easrly '70's) and took it to Africa, by the end of which it wasn't really worth using. My Nikons have been all over the world and bounced around all over the place without problems, dropped my 15mm in a salt water pool once and had to have it serviced.
    Read somewhere on here that someone had to throw their Canon EOS 1Ds Mark III out of a window to save it from confication, and retrieved it later, still in working order, but I suspect that like all delicate instruments, it had a lot to do with what it landed on rather than the throwing part.
    Both Nikon and Canon are tough instruments and are built for long and hard work.
    How many of us need 24Mpixels? Great for extracting small parts of the image, but you'd probably be better off getting closer to the subject in the first place, unless there is something preventing you, in which case use a longer lense, if you have one.
    More important is getting out there and taking the photos you want, if it needs 21Mp or 24Mp to obtain them, then go buy the apropriate camera and a large hard drive to accommodate the images.
    Has anyone done a shoot with all the different lenses/cameras (both SLR & compact) of the same subject, processed in Photoshop or a.n. other processing package, and reduced them to photo.net viewing size and compared the results to see what the differences are to the average viewer? It would be interesting.
     
  61. the processors, etc., just format and shuffle the data. If they affect the image quality at all, there's a problem
    This isn't true. The difference in output from different image processing and raw conversion algorithms can differ dramatically. The image processing and raw conversion is where the "look" of the image is determined; different processing is equivalent to having different films except that the spectral sensitivity cannot be changed without altering the filter array. With digital you can you can specify things mathematically instead of having to implement it in chemistry, the latter being far more difficult. There are things that you can't change without going to hardware, such as the spectral responses of the photosites, so in that sense different cameras may be needed e.g. for IR work etc. But one doesn't need different cameras for night, portrait, and landscape work.
     
  62. the processors, etc., just format and shuffle the data. If they affect the image quality at all, there's a problem
    This isn't true. The difference in output from different image processing and raw conversion algorithms can differ dramatically.
    You snipped the important part of what I said. Yes, the raw conversion affects image quality -- but when you're shooting raw, the raw conversion is not done by the camera. The whole point of raw format is that you're getting the data in its raw format -- no processing has been done to it yet. The raw conversion is then done (on your computer) by Capture NX, ACR, Bibble, DxO, or whatever raw converter you happen to prefer -- but the Expeed processor in the camera has nothing to do with it.
     
  63. I look at 4x5" view camera photos taken back in 1914, and how clear they are. Digital can't do that yet. I want 50+mp's in my camera so I can grab that much detail. I remember when 4mb of RAM in a computer was enough to run any application. Now 4gb is the standard. Same with digital cameras. 12mp is really the minimum, 24 is better, 48 is even better.
     
  64. " The whole point of raw format is that you're getting the data in its raw format -- no processing has been done to it yet."
    Actually you are wrong about this. All 'raw" images out of all cameras (that I know of: Canon, Nikon, etc.) do some degree of processing in camera on the path from the receiving CMOS or CCD to the media card (CF /SD). All of them.
    Two examples; Canon (possibly Nikon and the others as well) applies a default amout of sharpening to the data. Canon does this a bit more aggressively than Nikon however. And althoug hit is done by processors on the imaging CMOS "chip" itself Nikon's Active D-Lighting functions, is applied in camera as well.
    The only cameras that I know of that does not do any "pre-cooking" of the raw data are some of the medium format backs and the 4x5 BetterLight Scanning backs.
     
  65. "I look at 4x5" view camera photos taken back in 1914, and how clear they are. Digital can't do that yet. I want 50+mp's in my camera so I can grab that much detail."
    Digital can do it, you just need to think larger. You want somethign like the Leiica S2, or the other high megapixel medium format backs or one of the 4x5 BetterLight scanning back.
    Thsi starts to come down to format size issues. From the 1990s I've got 4"x5", medium format film and 25mm film , all low ISO transparencies, of the same subject shot nearly simultaneously and it is an easy guess as to which format has better detail rendition from deep shadow to highlight.
     
  66. Ellis,
    You may be right, but I'm still certain that more resolution will be demanded by photographers. It's no different than film being engineered to have finer grain and more detail. I have no plans to buy into a large and expensive medium format camera system. I can still shoot ISO 25 black and white through my Rolleiflex, but then I have to scan it and I have to rent the scanner as all I have is a Nikon Coolscan V for 35mm film, which does do a nice job.
    00Rlq1-96895784.jpg
     
  67. deleted
     
  68. Ellis,
    You need to update your information, a RAW file is a 2D array of 14-bit or 12-Bit intensity values, sharpening, color etc. are post demosaic concepts , you can't apply regular "sharpening" to an intensity level it makes no sense.
    http://en.wikipedia.org/wiki/Raw_image_format
    Also CMOS is a common name for an APS (Ative Pixel Sensor) fabricated in a traditional CMOS process and does not refer to a digital signal processing chip.
    http://en.wikipedia.org/wiki/Active_pixel_sensor
    DSP is only possible post quantization of analog signal and is performed by a DSP ASIC which is on a seperate PCB mounted behind or next to the sensor PCB.
    00Rlzd-96963784.jpg
     
  69. "Ellis,
    You need to update your information, a RAW file is a 2D array of 14-bit or 12-Bit intensity values, sharpening, color etc. are post demosaic concepts , you can't apply regular "sharpening" to an intensity level it makes no sense.

    This is the difference between theory and what happens in the real world. Note that Canon raw files are not .raw, they are .crw or .cr2 files. Canon has done somethign to them on their way from individual photo site to the recording media. Same thing with Nikon NEF, etc. Generically for they are known as "raw" files but in the real world , for the most part, they are not. Canon's application of some fairly aggressive initial "capture sharpening" to .cr2 files is pretty well known.
     
  70. Here's something that might interest you. I have no idea whether or not this is a valid comparison.

    http://masterchong.com/v3/sony-alpha/nikon-d3x-vs-sony-alpha-900-iso50-iso6400-100-cropped-comparison.html
     
  71. "This is the difference between theory and what happens in the real world. Note that Canon raw files are not .raw, they are .crw or .cr2 files."
    Ellis,
    Maybe you are not very familiar with computers but the extension of a file is just an arbitrary designation, it can be set independent of the content, you can just rename the *.CR2 or *.NEF to *.RAW if you like, the extension is just a tag so the appropriate application can identify the file. There is no such thing as theoretical RAW or real world RAW. If you are making such strange claims you need to provide a valid reference.
     
  72. ShunCheung

    ShunCheung Moderator

    This thread has gotten quite off topic. I am closing it for now.
     

Share This Page