Nikon sloppy, can we trust it ?

[andylynn]

<p>Maybe you're not looking at photos by good photographers?</p>

<p>But anyway. Here's how it works. Say Nikon sell 1,000,000 D800's. Of those, say (and I'm making up these numbers, they're not real numbers) 1,000 have a problem related to their AF systems, 10,000 of them are sold to people who use them with a lens that has a strong focus calibration error and they don't know it, and 39,000 of them are sold to people who are incompetent, or have lousy technique, but have a lot of money and think that buying the newest camera automatically makes their pictures better.</p>

<p>So 95% of those 1,000,000 D800's are being used to take photos that are in focus. Since the users are happily shooting, they're not starting forum threads called "My D800 has no focus problem!" because they didn't know it was expected.</p>

<p>0.1% of the cameras have an actual problem and if those people manage to get them to the shop they'll get repairs or replacements for no charge.</p>

<p>4.9% of the cameras are being used to take blurry shots, for reasons that have nothing to do with the camera. Only the users don't know this stuff well enough to identify the real source of the problem. A lot of them are mad at Nikon for selling them a lousy camera and post in forums about it (because mad people are vocal). So if 2% of the 50,000 people who are taking blurry shots that aren't caused by a camera problem are posting in DPReview and Flickr boards (and occasionally here, but hopefully there are people with helpful advice, even if that's "return your D800 and get a Powershot and photography lessons") it looks like Nikon has screwed up, big time, and shipped an expensive camera with broken AF.</p>

<p>The same thing happened with the D7000 and the iPhone 4. (Only that was a real thing - for people with an uncontrollable need to grip whatever they're holding with vise-like strength. Which explains why they don't have girlfriends and can't play golf, and so have nothing better to do than complain about their phones on the internet.)</p>

[bob_sunley]

<p>As far as costs , ok, a current Nikon F6, $2400 retail, give or take. Add 36,000 frames of film at $3-7 per roll, $5,000 average let say, add processing, scanning , another $2-3,000. You are now at $11,000.</p>

<p>$3,500 or so for the D800 is not so much money.</p>

<p> </p>

[wogears]

<blockquote>

<p>Most recently I was told by a D800 user to forget about buying one unless I upgrade all my lenses to new 'N' coated versions or I'm just wasting my time.<br>

</p>

</blockquote>

<p>Is it legal to shoot gearheads? Bet my 105 f2.4 P (AI converted) will make gorgeous shots on a D800. Bet my 35-70 f2.8 will do the same.</p>

[Andrew Garrard]

<p>Bob - if that was at my cost justification for people wanting to make sure their lenses could keep up with a D800... Something tells me the D800 might be out-selling the F6. I might even guess that it out-sold the F6 in the first week. And, for some reason, I doubt there are many people who put 36,000 frames through their F6s (F5s yes, but I'm not sure that many F6s are actually used commercially). Though, of course, if we're going to play "does my lens outresolve my film", we'd have been drum scanning all those rolls.<br />

<br />

I would suggest that not many people bought an F6 in order to get sharper photos; to put it another way, while I have a fairly low opinion of the world's population, I don't think there are that many stupid people with that much disposable income. I suspect quite a few D800 buyers bought them in order to get 36MP of useful image out of them (although what they do with it other than pixel peep at the sharpness or lack thereof is another matter), so realising that this investment also required - for best results - a lens (and technique, and possibly support) upgrade is significant. Not everyone just wants resolution. My 28-200 spends a lot of time on my D700, but I'm not expecting it to keep up with a D800E (although it's only about 1.7x sharper, to be fair...) However, I don't care - sometimes there are priorities over using high-end primes for every shot. When I really want 36MP, I'll worry about technique and equipment. Going out and buying a 28-300 would do a better job of making me poor than helping, after all.<br />

<br />

I suspect a lot of F6s sit on shelves, a lot of film never gets scanned, a lot of film that <i>is</i> scanned is scanned in a consumer flat bed scanner, and a lot of people quite rightly believe that extinction detail is all very well, but if you've got to pick through grain the size of billiard balls to see it, maybe the limits of the lens don't matter much. I doubt many people are shooting with an F6 for absolute technical quality - a Mamiya 6/7 or a Hasselblad, maybe; a 5x4 or larger definitely, but 135 is a convenience format and in this world of mass-market lenses most would be "good enough". With a D800, you can see the limits both of the lens and of the autofocus system - and of technique.<br />

<br />

You're quite right that it probably doesn't really matter, but any problem that's visible, people will complain about. (Though you could also argue that a D700 is plenty of camera for most people, and that Nikon would do a lot of people favours by lanching a D600.)<br />

<br />

Incidentally, the F6 looks like a lovely camera, and I'd love to have one... but not for the going rate. I can't afford to use it as I use a DSLR (by Bob's maths). I settled for an F5. But I'd love an F6 for the handling - not because it's going to make anything sharper.</p>

[bob_sunley]

<p>I only mentioned the F6 as it is the only "new" pro grade film camera available from Nikon. Otherwise it would be a cost comparison between a used film body and the new D800.<br>

Except for production costs of prints, digital is close to being a hands down winner now for the hobbyist. RA4 prints done in a home darkroom from film are much lower in cost than having a lab or buying the hardware/paper/ink to do a nice 11x14 or 16x20 print.</p>

[Andrew Garrard]

<p>I think we may have crossed wires, and I'm afraid I'm under-slept enough that I've forgotten what point I'd been making. I think I was just trying to say that people are spending a lot of money on a D800 in order to get resolution, which means that spending some money on upgrading the lenses to get more out of it makes sense. Buying a D3200 and not being able to get the whole 24MP is less annoying/wasteful because it's so much cheaper. In the film world, I'd be annoyed if I bought a Mamiya 7 for resolution and had lenses that couldn't get the same overall resolution as I could have achieved with 35mm. I'd be annoyed if a lens used on 5x4 could only resolve the detail from a decent 35mm lens - e.g. if it could only do about 35 lp/mm. Unless I were upgrading from APS, I'd have relatively low expectations of a 35mm film system in terms of resolution. It's not like an F6 produces sharper images than an F75, so the justification for high end lenses isn't there - and no (common) film is really expensive enough to justify it. But then I don't work in the film industry. :-)</p>

[zack_zoll]

<blockquote>

<p ><a name="00ackb"></a><a href="../photodb/user?user_id=112337">Les Berkley</a> <a href="../member-status-icons"><img title="Subscriber" src="../v3graphics/member-status-icons/sub10plus.gif" alt="" /><img title="Frequent poster" src="../v3graphics/member-status-icons/3rolls.gif" alt="" /></a>, Jul 17, 2012; 09:44 a.m.</p>

</blockquote>

 

<blockquote>

<p><em>Most recently I was told by a D800 user to forget about buying one unless I upgrade all my lenses to new 'N' coated versions or I'm just wasting my time</em>.<br>

Is it legal to shoot gearheads? Bet my 105 f2.4 P (AI converted) will make gorgeous shots on a D800. Bet my 35-70 f2.8 will do the same.</p>

</blockquote>

<p>You're half right. The D800 is much more demanding on lenses, and your 35-70 will 'magically' take worse photos, unless you stop it down pretty far. I don't know about your 105 though. I do know that the 105 f/1.8 is pretty much worthless on the D7000 and the NEX-7 unless you stop down pretty far, while the humble 135 f/2.8 is acceptable wide open on those cameras, and excellent at f/4 and below.</p>

<p>But the bit about needing N lenses is bunk. There are a number of lenses that outresolved the capture technology that they were designed to be used for, and these will still take excellent photos, regardless of age. Zeiss glass comes to mind, along with (to a lesser extent) several older manual Nikon lenses, although flare are haze will likely be an issue wide open, and you'll REALLY want a lens hood.</p>

<p>You can even try putting a Hoya SMC filter or similar in front of an older lens to help mitigate the issues of older coatings. It's not supposed to work, but it does - just barely. It won't turn a bad lens into a good one, but if you're on the fence about whether or not the lens works as well as you like with the camera, it's just enough of a difference to make you keep it. My 50 f/1.4 AIS had some minor CA issues with newer bodies, and buying the filter helped. It still happens like crazy when the subject is backlight, but other than that I don't get CA anymore unless the lens is wide open and the subject is super contasty.</p>

<p>But like I said - it's an extremely minor difference. I had to pixel peep to see the CA in the first place.</p>

 

[Andrew Garrard]

<p>Zack: I'm not sure I understand how the extra filter could help with flare, haze or CA. If it wasn't reflecting light around itself, it still wouldn't stop the lens's internal reflections. Are we talking about UV?<br />

<br />

(Not disputing your findings, just trying to understand them.)</p>

[Peter_in_PA]

<blockquote>

<p>The D800 is much more demanding on lenses, and your 35-70 will 'magically' take worse photos, unless you stop it down pretty far.</p>

</blockquote>

 

<ul>

<li>It all depends on how you USE your photos. If you shoot 8 x 10s or so, you'll be fine... and...</li>

<li>If you stop down too far, you will get diffraction, which is a whole other beast, and the smaller your photosites, the worse it is... at the pixel level. So the D800 is more prone to diffraction at the pixel level than, say, a D700.</li>

</ul>

<p>The thing that all these discussions forget is that the "softness of old lens" issue with something like the D800 is only apparent at THE PIXEL LEVEL! If all you are doing is pixel-peeping, then by all means, it's a problem, but if you are taking great photos, printing them, and enjoying them, probably not.</p>

[steve m smith]

<p>How can a lens suddenly be worse when used on a higher resolution camera? That is nonsense.</p>

[Peter_in_PA]

<blockquote>

<p>How can a lens suddenly be worse when used on a higher resolution camera? That is nonsense.</p>

</blockquote>

<p>It's only possible if you measure quality by the pixel and not by the image. In this day and age, many do.</p>

[cyrus_procter]

<p>Well for what ever its worth, I have an old 15mm F/5.6 AI lens. Super limited production and in its day (late 1970s) it was a truly amazing lens. Today its blown away in every aspect by lenses like the 14mm F/2.8 or the 14-24mm F/2.8 except one, and that is distortion. On paper the 15mm F/5.6 has less distortion than Nikon's latest 50mm F/1.4G. I used it occasionally on the D3s but felt it was too soft. On my D800 at least on monitor\11x14 sizes, I actually feel like its sharper. Sure if I blow it up 100% its not that sharp and is easily softer than the aforementioned modern wide angle lenses, but on medium to small sizes its actually sharper.</p>

<p>Somethings you have to take with a grain of salt, like Canon's old 24-70 being sharper than Nikon's 24-70. Of the 3 copies of Canon's 24-70 I've gotten to use my Nikon 24-70 would be sharper at F/2.8 with petroleum jelly smeared on the front than Canon's old 24-70 at F/8 (and I've used Canon's old 24-70 on the 5D mkIII and gazillion times on the mkII). Well ok, I exaggerate, but the point is, when you hear something like that, you know there is something wrong with the reviewers, or else they hit that .001% of lenses that really is soft.</p>

[Andrew Garrard]

<blockquote><blockquote>How can a lens suddenly be worse when used on a higher resolution camera? That is nonsense.</blockquote>

It's only possible if you measure quality by the pixel and not by the image. In this day and age, many do.</blockquote>

 

<p>That's not <i>quite</i> true: if a lens is sharp across some of the frame but not in other regions, the overall effect can sometimes look worse than if it was uniformly softer. As such, a lens that's always "sharp" on a D700 and only selectively sharper on the D800 can result in the D800 image seeming inferior. But it's subtle, probably unusual, and you could always blur the D800 image down to D700 resolution to solve it.<br />

<br />

Skyler: Glad I wasn't imagining the Canon 24-70 performance. I might drop AP an email...</p>

[zack_zoll]

<blockquote>

<p ><a name="00acnA"></a><a href="../photodb/user?user_id=4303235">Andrew Garrard</a> <a href="../member-status-icons"><img title="Frequent poster" src="../v3graphics/member-status-icons/3rolls.gif" alt="" /></a>, Jul 17, 2012; 01:32 p.m.</p>

</blockquote>

 

<blockquote>

<p>Zack: I'm not sure I understand how the extra filter could help with flare, haze or CA. If it wasn't reflecting light around itself, it still wouldn't stop the lens's internal reflections. Are we talking about UV?<br />(Not disputing your findings, just trying to understand them.)</p>

</blockquote>

<p>Andrew- current high-end lens coating technology (like the N coatings) does an amazingly good job of reducing flare, CA, and general haziniess. Old lenses may be very sharp, but will lack the newer coatings. Between the smaller photosites of newer sensors, and the thinner section of the sensor that is actually sensitive to light as compared to a piece of film, things like CA will crop up more frequently. Remember that CA is caused by different colour wavelengths being out of alignment with each other.</p>

<p>Introducing a SMC UV filter (sorry I forgot to mention the UV part) helps a little bit; it's like having a single N coated element in the lens. It doesn't change much, because the rest of the lens still has older coatings, including the front and rear elements, which matter the most. It also does absolutely nothing to help internal reflections.</p>

<p>Like I said, it's barely measurable, and I had to shoot a lot of photos with and without the filter to see much difference. But if you're <em>almost</em> happy with an old lens, it might be worth a try. It fixed my Nikon AIS lens, but it did basically nothing for my Minolta 58 f/1.2, which has MUCH worse CA wide open.</p>

<p>As far as lenses not becoming 'less sharp' on higher resolution cameras ... well I'm sorry if you don't believe me, but that's just the way that optics work. ANY optical device has a maximum resolution, be it a lens, a scope, or a pair or glasses. If (for the sake of argument) a lens has a maximum IQ that is in line with a 10 MP sensor (say it's a CCTV lens), putting that lens on a 40 MP camera will still yield the same amount of information as putting it on a 10 MP camera. There will be four times as many pixels, which means that a larger print may show less pixellization. But the lines on the print will not be much sharper than on a 10 MP sensor, because the lens can't provide any more detail than that.</p>

<p>Obviously the lens doesn't physically change, so it doesn't technically become a 'bad lens.' But it does become a bad lens to use with that 40 MP camera. And as time goes by, and sensor resolution and the needs of photographers change - and lens technology improves -we will begin to see it as a bad lens.</p>

<p>If you've ever used old prewar cameras, you know what I'm talking about. If you've had the good fortune to stumble on old, preserved film stock, then you're definitely on my page. Those lenses were awesome <em>then</em>, because they provided more detail than the film stock of the time could resolve. But as film emulsion improved and grain sizes shrunk, it became apparent that some of them were not quite as sharp as originally thought.</p>

<p>That has a lot to do with why a 60+ year old LF Zeiss lens can be had for cheap, while a new one with a similar lens design cannot.</p>

 

[ant_nio_gomes]

<p>Congratulations Maurice,<br>

You don't have this camera, not even any other Nikon model, but you come here with a lot of "information" on the problems, failures and manufacturer's wrong behavior to criticize Nikon and make a statement on your doubts about how trusty they can be for you...as a potential future customer, that we may conclude will buy another brand.<br>

You stated all this, caused 4 pages of comments even if you already vanished. Great achievement that grants the right to a big applause.</p>

[ilkka_nissila]

<p><em>The D800 is much more demanding on lenses, and your 35-70 will 'magically' take worse photos, unless you stop it down pretty far.</em></p>

<p>Could you demonstrate this with pictures?</p>

<p>Theory suggests that all lenses should resolve more detail in all parts of the image when a higher resolution sensor is used (unless there is some kind of incompatibility between the lens and sensor optics). If you're unhappy with having 80% of the image with much higher sharpness and 20% of the outer parts only somewhat sharper than the D700 image, you can blur the image evenly to come up with an image that has homogeneous sharpness across the frame (although I doubt any actual user of an D800 would do it after seeing the images). If you do this right, all parts of the blurred image will still show a bit better detail than the corresponding parts of the D700 image (assuming the corresponding D700 image has even sharpness across the frame).</p>

<p>I understand the psychological effect of having large files which don't look sharp at the actual pixel level, but that's just life, I would advice people to get over it; future sensors will only have more pixels, not fewer. It is a good thing that the sensor has enough pixels to show the true character of the lens; otherwise one might spend a lot of money on lenses and never see the full benefits of the investment (as the investment in lenses is usually much greater than the investment in cameras, you can see there is a problem if the sensor just shows ten percent of the capability of the lens). AFAIK many high end lenses show 100lpmm on Tech Pan, so to resolve all the detail that a good lens render at its optimal aperture, using a Bayer sensor you'd need something like 300MP. </p>

<p>The real drawback of the D800's high resolution is that it consumes storage and computing resources in an amazing way and at least for me the real cost of the camera is . However, the image quality is phenomenal. Although lenses can resolve more detail than that, AF technology isn't keeping up, and in practice I don't make all that many large prints, so I would be happy to settle with something around 24MP in the long term for practical reasons and focus my computer purchases on making post-production faster. However this doesn't appear to be what the future will hold, unless I go for the "fast" models of the DSLR lineup (i.e. D4) which are priced quite high up there.</p>

[Andrew Garrard]

<p>Zack: I can see that a (functioning) UV filter might affect chromatic aberrations if the camera is picking up UV. That does imply that the camera has an ineffectual internal UV filter, though. I'm just confused as to how it can have any other effect: the coatings on lenses are designed to increase transmission and limit internal reflections; putting anything on the front of the lens shouldn't have any effect on this. I'm happy to give it a try at some point, though - just because I'm confused doesn't mean that something I don't understand might be going on.<br />

<br />

Regarding the "new lenses for a D800" arguments:<br />

1) For the most part, the question is "is your lens good enough to make the most of a D800 sensor - and if not, why did you buy a D800?" In other words, I suspect most people buying a D800 have done so for resolution reasons, and it's worth pointing out that they're (mostly, I accept Ilkka's point) wasting their money (and getting other D800-related compromises) if their lenses can only keep up with a D700. Or at least, that a lens which appears perfectly sharp on the D700 (because it resolves beyond the sensor) won't appear so on the D800. In the grand scheme of things, there's nothing wrong with not being able to resolve to the limits of the D800 sensor, but many people are used to being able to do so, and paid a premium assuming that this would be the case. I anticipate a number of D3200 owners complaining about their lenses as well, but probably fewer will have a back catalogue of lenses compared with D800 owners.<br />

2) The image as a whole will not be less sharp with a lens used on a D800 than on a D700. Nonetheless, people tend to buy D800s because they either want to do bigger prints or more cropping than is possible with other cameras. If they're doing this, their lenses may now look soft. If the image is the same size as before, it won't - but then why get a D800?<br />

3) I claim that an image which is selectively sharp can, occasionally, appear less appealing than an image which is universally soft. This isn't going to stop me buying a D800 and probably using it with some lenses that can't keep up, but I wanted to make the point that it's an effect that can happen.<br />

<br />

Ilkka: I would not be surprised to find a 24MP (or 16MP?) D600 launched soon. A 24MP version would be more of a direct competitor to the 5D3, especially if it can hit 6fps at full speed. A 16MP version would be more of a direct successor to the D700. I doubt it would best 8fps (because it would compete with the D4); a 24MP version that hit 7-8fps might compete well against the 5D3, but I suspect would also impact D800 sales, so I doubt they'd be able to aim such a thing at the "low end FX" segment and undercut the D800 much. That said, if Nikon were to get rid of any remaining D3s sensors by selling a "D700s" special edition, I'd... well, really wish I could afford to get one as well as a D800.<br />

<br />

Not that, of course, I know anything (in general, but especially about Nikon's plans). But I'd wait a little while before deciding that the D800 is a definitive guide to "the future". I'm curious as to whether Canon will feel the need to respond to the D800 with one of their high pixel count sensor prototypes (perhaps revamping the 1Ds) though. I doubt they really care, if it's true that the DX sensors make all the money; the D3200 might have done more to set the cat amongst the pigeons, but it's not like Sony haven't been using this sensor for a while (and Samsung have a 20MP one), and the D3200's sensor is low-light compromised compared with the D5100's in a way that the D800 isn't.</p>

[ilkka_nissila]

<p>Andrew, I think people will buy the D800 because it gives improved image quality at any size print beyond the postcard. Pixel count and print size are not related in any simple way - I make prints at any size from any original depending on application, not based on the pixel count of the original. It doesn't matter if you use a mid level or a great lens; the results are improved in each case. It's not like the D800 turns Nikkors into single-element Lensbabies with distinctly unsharp outer parts - this isn't how it works. The D700 massively underresolves most lenses; this was always known as people scanned 35mm film (which is a much lower quality medium that FX digital) typically at 4000 ppi (20MP; no bayer) and the Minolta 5400 ppi scanners produced even better sharpness (40+MP) that was obvious in magazine reproduction in a double page spread (but there were problems with film flatness). Nikon's decision to 12MP with their first two FX cameras was most likely because that's the only FX sensor they had available at that time that was finished, and it was developed for sports/action mostly. It's not because lenses couldn't outresolve it - almost all good lenses basically saturate the D700 sensor with detail at most apertures. Yet even those that do not appear to be pixel-sharp on D700 are still much improved with D3X and D800. There is no such thing as lenses that can't "keep up."</p>

[Andrew Garrard]

<p>I'm expecting people to buy the D800 for image quality; I just think it's reasonable to say that, if a customer is used to pixel-level sharpness on a D700, if they're expecting the same level of pixel quality from the D800, they might need to check which lenses they own.<br />

<br />

I'm actually surprised how large the difference is - the D800 has only 1.7x more pixels in each direction than a D700, yet it does seem to be showing up lenses. It's possible that a lot of lenses have been optimised for higher contrast at lower resolutions, and absolute resolving power has suffered - the D800 is "beyond the cliff". (I've yet to get that book I've been promising myself on lens design, but I understand that these are conflicting requirements.) Of course, from the perspective of looking at the image, there's no problem with such a contrasty design - it's only if the D800 is being used with the intent of recording more detail; but then, this is presumably the point.<br />

<br />

As for lenses that don't keep up with the D700... well, the original 50mm f/1.8 (and I believe f/1.4) aren't very sharp until you stop them down. The same is true of the corners of the magical 50mm f/1.4 Sigma, and the 85mm f/1.4 AF-D. The mk1 70-200 VR didn't have sharp corners at 200mm. At least my sample of the 150-500 Sigma is visibly soft at the 500mm end unless stopped down a lot. The Sigma 20mm f/1.8 doesn't have anything you'd call resolution away from the centre. With a D800, I'm expecting more lenses to show their limits (certainly the 14-24; I'm not convinced by the sample images from the 24-70 either), but while the D700 could be called "forgiving", that doesn't mean that <i>every</i> lens could outresolve it.<br />

<br />

None of this makes those lenses useless, but if I were buying a D800 with a view to getting resolution, and I'd previously bought less sharp lenses in the knowledge that any lack of sharpness wouldn't show up on my D700 (e.g. the 80-200 I bought recently in place of a 70-200), a heads up that a new lens might make a significant difference is valid.</p>

[ilkka_nissila]

<p>Andrew, those lenses are <em>designed</em> to have spherical aberration for the purposes of controlled softness for portraits and for the out-of-focus areas (although you can blur skin with photoshop often to atrociously cartoonish effect as seen in many fashion magazines, with film purpose-made lenses were very useful, making people look beautiful in adverse lighting conditions without labourous retouching). Spherical aberration increases softness towards the corners; this is the tradeoff. It is not a lens flaw, it's the way the designers made it. The 70-200 Mk II has improved corner sharpness but it gives a really hard rendition of faces, with deep shadows (which is why I a lot of the time use tele primes in hard light) and the background rendition is distinctly less smooth at longer distances. Nikon altered their design because of complaints by landscape photographers but in doing so they made it an inferior portrait lens (IMO). Sometimes I think I should have kept both. The transition from 85/1.4 D to G went better though again I have seen harsher bokeh in some circumstances with the latter while not the former, but overall the G is a much improved lens as it gives an excellent image already at f/1.4 and has the focus accuracy to back it up.</p>

<p>None of this however makes these older lenses any worse to use on a D800; in fact I can easily demonstrate that the 50/1.8D (which is the only one on your list of older primes that I have kept) is much better on the D800 than on the D700.</p>

<p><em>the D800 is "beyond the cliff".</em></p>

<p>There is no "cliff." Lens MTF and sensor MTF both drop very g r a d u a l l y. This is why even though defects may show on a D700 the D800 is able to lift them to a great deal. Even at web sizes, images from the D700 show noticeable effects of the heavy-handed antialias filter that attenuates these low frequencies while the filters from the D3X and D800 do not. I can't apply as much sharpening to D3X or D800 Wednesday posts that I used to, with the D700 and D3 files - it would just look harsh and oversharpened.</p>

<p><em>I'd previously bought less sharp lenses in the knowledge that any lack of sharpness wouldn't show up on my D700</em></p>

<p>Now that's just bad planning. You have to look at present needs for sure, but also make purchases that will stand the test of time. Even before I bought the D3 in January 2008 I had been stocking up on lenses which are known to give excellent sharpness across the full frame (such as 35/2 ZF), since I knew the D3 is really the D3H and the X was coming and that path would be the future for 35mm digital. In 10 years the high speed cameras are also high res (certainly 24MP, possibly 36MP or more) since computational and storage limitations will have been moved up. For my practical needs 24MP is sufficient and 36MP is an unnecessary burden for my PC but in 5-10 years things will look very different. Even the small jump from 24MP to 36MP has made some of my less good lenses (135/2 wide open, 70-200II+1.4X) a bit better, which is great. The D800 saves money that I would otherwise have to put into more expensive lenses to reach a given level of performance. In fact the 70-200+1.4X has been my biggest surprise so far, getting good results on a zoom with TC was something I hadn't expected. Yes, at f/5.6 the results are more cluttered than with 300/2.8 at f/4 or f/2.8 but the portability and flexibility are great assets for concert photography. I still want the 300/2.8 but I might just be able to wait a few years for a 300/4 VR, which would be a much better match for my needs and suitable for travel.</p>

<p>I have made my share of bad purchases - I didn't much care for the 25/2.8 ZF, and the 24-120/4 I bought even though I knew pretty much what I was getting - I just wanted to give it a chance. I'm happier without, even though it's on Nikon's list of recommended lenses for D800... others will have a different take.</p>

[Caledonia]

<p>I think a lot of people on forumns just like stirring it, you could become paranoid very easily if you believed everything you read. Having had the d800 now for about 6 weeks I can honestly say it is a Landscape photographers dream camera,the 30x20 inch prints I have recently had done are amazing.</p>

[eric_arnold]

<p>just stumbled across this discussion, which, at least to me, is typical of Internet forums these days. some troll makes a highly opinionated post that has to be taken with a grain of salt because it's all just speculation about a camera he doesnt have, then everyone falls over themselves with analysis paralysis and seemingly contradictory assertions. if we stick to the facts, what do we find?</p>

<ul>

<li>the d800 is the highest-resolution camera nikon has ever made</li>

<li>there are some known QC issues - just like with any new product</li>

<li>some lenses work well with it; others dont</li>

<li>it produces extremely large files</li>

</ul>

<p>everything else is just grist for the mill. beyond that, what can be surmised?</p>

<ul>

<li>the d800 is not for everyone. obviously, there will be a lower-res FX sensor, sooner rather than later.</li>

</ul>

<p>but c'mon, it's just plain dumb to call nikon "sloppy" and question its future as a reliable brand, especially when the OP can't even muster enough bravery to respond to a single point of feedback. look at complaints about fuji's X100, or canon's 5diii, and you can either conclude that hi-end camera makers intentionally put out unfinished products, or hi-end camera buyers are whiny lil' brats.</p>

[Andrew Garrard]

<p>(Apologies for the delayed reply.)</p>

 

<blockquote>Andrew, those lenses are designed to have spherical aberration for the purposes of controlled softness for portraits and for the out-of-focus areas</blockquote>

 

<p>That may be true of the 85 f/1.4 (certainly of the bokeh), though the smearing away from the centre zone is pretty ugly to my eyes. I'm sure it was a consideration with the 50mm lenses, but the fact that there isn't a sharp alternative (at wide apertures) and that the 50mm f/1.8 AF-D's bokeh is so ugly makes me think that the primary concern was making a cheap tessar derivative. The Sigma 50mm is clearly designed for bokeh, but its sharpness is also targetted at DX shooting - lens design often seems to include a trade-off with a bounds to the optimal image circle, and this is also true of the Mk1 70-200. The 150-500 (mine, anyway) and 20mm Sigmas are just soft, with no excuse. Soft focus with a genuine SF lens (like Canon's) or with deliberate LoCA (like the DC lenses) is understandable, but I really think that some of the more conventional lenses still had sharpness as a primary requirement. I'm not assuming that all the lenses recommended for the D800 have significantly worse bokeh than their predecessors - though I buy your arguments that Nikon may have de-prioritized bokeh in the 70-200 and 85mm updates.</p>

 

<blockquote>(although you can blur skin with photoshop often to atrociously cartoonish effect as seen in many fashion magazines, with film purpose-made lenses were very useful, making people look beautiful in adverse lighting conditions without labourous retouching).</blockquote>

 

<p>Each to his own. I would prefer to have a sharp image and control over softening it (admittedly, enough control to avoid the "cartoonish" look) than try to revert lens aberrations.</p>

 

<blockquote>Spherical aberration increases softness towards the corners; this is the tradeoff. It is not a lens flaw, it's the way the designers made it. The 70-200 Mk II has improved corner sharpness but it gives a really hard rendition of faces, with deep shadows (which is why I a lot of the time use tele primes in hard light) and the background rendition is distinctly less smooth at longer distances.</blockquote>

 

<p>I buy that the bokeh of the Mk II suffered. Still, I would suggest that the Mk I was designed for DX sensors - for which the corner performance is irrelevant. I don't think Nikon deliberately designed this lens with soft corners.</p>

 

<blockquote>Nikon altered their design because of complaints by landscape photographers but in doing so they made it an inferior portrait lens (IMO). Sometimes I think I should have kept both.</blockquote>

 

<p>I hear you, although I would suggest that it's not just landscape photographers who won't put up with a smeared mess in the edges of the frames. But then I only bought an 80-200 as an afterthought for when I want focal length and DoF control - I dismissed both 70-200s on their cost-to-LoCA ratio (and got a 200 f/2, which seems to be sharp and have good bokeh).</p>

 

<blockquote>None of this however makes these older lenses any worse to use on a D800; in fact I can easily demonstrate that the 50/1.8D (which is the only one on your list of older primes that I have kept) is much better on the D800 than on the D700.</blockquote>

 

<p>Oh, I quite agree. I also have a 50 f/1.8D, and I'm sure I'll use it on my (impending) D800. Stopped down, I expect it to do very well. I was just including it as a lens that doesn't outresolve the sensor of a D700 used wide open (at least off-centre).</p>

 

<blockquote><blockquote>the D800 is "beyond the cliff".</blockquote>

 

There is no "cliff." Lens MTF and sensor MTF both drop very g r a d u a l l y. This is why even though defects may show on a D700 the D800 is able to lift them to a great deal. Even at web sizes, images from the D700 show noticeable effects of the heavy-handed antialias filter that attenuates these low frequencies while the filters from the D3X and D800 do not. I can't apply as much sharpening to D3X or D800 Wednesday posts that I used to, with the D700 and D3 files - it would just look harsh and oversharpened.</blockquote>

 

<p>I should rephrase. My understanding - and this could be incorrect, and I'm happy to learn - is that, while lens MTF should monotonically decrease as frequency increases, different lens designs can alter the shape of the MTF-to-frequency curve. In addition, I understand that there's a trade-off between extending the tail of the curve (so very high frequencies still resolve with reasonable contrast) and increasing the contrast at lower frequencies. If this is true, lenses may be designed to have more apparent detail (because reasonably high frequencies are resolved with good contrast) despite not actually recording as much detail as apparently softer lenses. By "cliff", I meant, perhaps, a "bulge" in the MTF-to-frequency curve beyond which contrast might fall off rapidly (but I agree, not instantly).<br />

<br />

Because there's a limit to the frequencies that a DSLR sensor can resolve, a lens which resolves at high contrast up to that frequency may be a better option for the sensor than a lens which trades lower frequency contrast for higher contrast at frequencies beyond the sensor's resolution. For this reason, I can imagine that lens A might be better than lens B on a D700, but the situation would reverse on a D800.<br />

<br />

Or I could be wrong. (And I agree about the D700 low pass filter - I'm expecting a considerable workflow shift with my D800E.)</p>

 

<blockquote><blockquote>I'd previously bought less sharp lenses in the knowledge that any lack of sharpness wouldn't show up on my D700</blockquote>

 

Now that's just bad planning. You have to look at present needs for sure, but also make purchases that will stand the test of time.</blockquote>

 

<p>It's a trade off. I bought an 80-200 just before my wedding (for a friend to use with my D700); since owning any zoom in this range is a bit of an afterthought to me (I already have good primes covering the range), I decided that a 70-200 just wasn't worth it, having compared them. The result may or may not be acceptable on a D800 (there seem to be some happy people, but the newer lens is clearly sharper, even on a D700); I'm sure the difference between them would be more obvious on the D800. Had I expected such a resolution jump, I <i>may</i> have chosen differently. Will my 28-200 hold up? I'm not convinced, but that lens has done me a lot of good service on the D700. I'm highly unlikely to keep my 135 DC or 150-500, but then I've been meaning to get rid of them for years anyway. But there are plenty of reasons for a trade-off - in addition to the apparent quality difference I referred to above, size and money are really significant. So far I've not particularly minded not having a Zeiss 21mm (though I do have a 35mm for my Pentax 645); with the advent of the D800, I'm no longer after a cheap 20mm f/4 and the Zeiss is looking tempting.</p>

 

<blockquote>I have made my share of bad purchases - I didn't much care for the 25/2.8 ZF, and the 24-120/4 I bought even though I knew pretty much what I was getting - I just wanted to give it a chance. I'm happier without, even though it's on Nikon's list of recommended lenses for D800... others will have a different take.</blockquote>

 

<p>It happens. I'm just expecting more of my selection to be relegated to "last resort" on the D800 than with my D700. I'm sure I'll have fun finding out (hopefully not on any shots that are important).</p>

[Andrew Garrard]

<p>Eric: I find that where conversations on this forum start is rarely where they end. Whether or not the original poster was expressing genuine concern - and whether it was just trolling - I've still been interested in the responses, especially as we wander off topic. I'd prefer we did this than just bite the head off anyone who posts something we don't agree with - if we're going to have a flame thread, at least let's make it useful.</p>

[eric_arnold]

<blockquote>

<p>The Sigma 50mm is clearly designed for bokeh, but its sharpness is also targetted at DX shooting</p>

</blockquote>

<p>andrew, do you own this lens? and have you used it on FX? i have on both counts and it's sharper at f/2 than the 24-70 is at 2.8. so i'm going to have to respectfully disagree with your analysis here. it's just a good lens, period, on both FX and DX, and aimed at users who will be using it at open apertures. i prefer it on FX, because the shallow DoF is even shallower. i dont think i've ever shot it narrower than 5.6 and only then on rare occasions.</p>

<blockquote>

<p>Whether or not the original poster was expressing genuine concern - and whether it was just trolling -</p>

</blockquote>

<p>i think maurice was genuinely concerned -- about his opinion. after all, he's the same guy who insisted no wedding shooter should do a gig without the 24-70, even if they shoot DX, awhile back. regardless -- and this should be the point-- there appears to be genuine concern about some faulty d800's. the problem is that blurting speculative assumptions on an internet forum without concrete context falls into the "chicken little syndrome" category. nikon's response to the issue may be less than desired, but it also may be the correct one from a business standpoint. no one knows for sure how many cameras are affected, so proclaiming the sky is falling when its only a portion of the sky may not help the situation. obviously, nikon doesnt want widespread panic to set in, and neither should we. there's an issue with SOME cameras, but not all. i'd much rather read a calm, reasoned, and informed analysis from thom hogan than the ravings of a madman regurgitating 2nd, 3rd and 4rd hand info.</p>