Pages: [1] 2 3   Go Down

Author Topic: Nikon D3x review from a landscape shooter's perspective  (Read 25396 times)

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1044
Nikon D3x review from a landscape shooter's perspective
« on: February 21, 2009, 12:49:52 am »

What is a D3x? It is a large, heavy, expensive "35mm" DSLR - as a matter of fact, it is tied with three Canons (none of which ever got the pricing flak that the D3x got) for the most expensive small-format DSLR ever mass-marketed (numerous Kodaks and a few other cameras marketed primarily to institutions were much more expensive). Unfortunately, that is where a lot of the conversation stops - people see the high price and the 24x36 mm image sensor, and file it in the undesirable mental category of "very expensive 35mm DSLR". In the digital age, however, sensor size is a lot less important than what a camera's capabilities are. I just returned from a landscape shoot with my D3x to Acadia National Park on the Maine coast, a location I have worked many times in a variety of conditions, with a variety of equipment, and have returned convinced that the proper way to evaluate the D3x (and perhaps some other very high resolution DSLRs) is as a MEDIUM-FORMAT camera. It provided the finest images I have ever made in that location (including film up to 6x9 cm, and a wide variety of digital cameras). I used it as if it was a 6x9 camera - locked down on a tripod at low ISO, and was rewarded with results that exceeded anything I've ever gotten from medium format film. It also has the versatility to shoot handheld when needed, and provide results that, while not equal to its own tripod-mounted results, exceed anything I've ever gotten from handholding any other camera. It has replaced medium format film for me, and, for the first time, I look at the final prints and there's nothing missing - I don't say "digital was nice for this aspect, but I wish I was shooting 120 for this other reason". The only film I've ever handled that provides more detail is 4x5 inch - and that does, even with a modest scanner - the D3x can take the lead in overall image quality on certain scenes if its extended dynamic range makes up for the 4x5 film's increased resolution, though. For my type of work (fairly traditional landscape, but in color), the D3x has replaced medium-format film (indirectly, I went through other digital solutions I was never quite satisfied with first), and caused me NOT to move to MF digital. It is a medium-format camera as far as my use of it is concerned, but it is also capable of being a 35mm camera when it needs to be.

      Different sensor sizes have blurred distinctions between formats - is a camera with an APS-C sensor, that otherwise handles like a 35mm SLR, and that would typically be used for 35mm applications, really in a different class from what we used to call 35mm cameras? A Nikon D300 or a Canon 50D appeals to the same group of photographers who would have previously bought a range of enthusiast 35mm cameras from the Nikon FM2 and Canon A1 to the N90 and EOS A2 series. Does the fact that their image sensor isn't the same size as 35mm film really mean they are a different kind of camera? Similarly, a D40 or the Rebel line are in the N65 or film Rebel class by any standard. The APS-C DSLRs are really "the new 35mm" in a very real sense, as they cover the same applications, appeal to the same users, and handle similarly. Some of the jobs that used to be the domain of Nikon's F series and Canon's F1 are now handled by APS-C cameras (especially the D300, popular for its quick AF), some by Canon's EOS-1D series with their intermediate sensor size, and some by the Nikon D3 and D700 with 24x36mm sensors.
     At the same time that cameras with smaller than 35mm sensors have taken a lot of 35mm jobs, with better than 35mm quality, some of the 24x36 mm cameras, which coincidentally have image sensors the size of 35mm film, have been designed for work that is well outside of the range of 35mm film. The first attempt at this type of camera was the EOS 1Ds, which probably provided better image quality than even slow 35 mm film given the most careful processing and printing. By the second generation, Canon's EOS 1Ds mkII and 5D, these cameras were clearly superior to any 35mm film under most conditions, and were often used where 645 film was the previous choice, generally offering an image quality upgrade even over most 645 films in many conditions. The third generation of ultra high resolution 24x36 mm DSLRs (EOS 1Ds mkIII, 5D mk II, Alpha 900 and D3x) are an upgrade in image quality from medium format film, even from formats larger than 645, in most conditions and types of photography.
     One possibly fair way to compare cameras is based on intended application - small format being built either for casual photography, speed or low light (or some combination) - all applications at which 35mm film excelled. Medium format cameras were (primarily, ignoring the Holga and its ilk) built for very high detail and resolution, in a package that was still reasonably convenient and could be handheld if needed (although designed to be used on a tripod much of the time). Most large format cameras were built for the ultimate in resolution, plus having camera movements to work with, at a huge cost in convenience. By this standard, the D3 is a small-format camera, while the D3x is medium format, despite having the same sensor dimensions and even the same body, due to their divergent sensor design philosophies. Most medium format digital backs count as medium format, but the highest resolution backs are clearly pushing large format territory, especially if equipped with a tilt-shift lens or adapter.
     The highest-resolution 24x36 mm DSLRs (including, but perhaps not exclusively, the D3x) are capable of printing as large as 24x36 inches, even from very high detail subjects - something that 35mm film could never do. The D3x, in particular, also lacks a couple of traditional 35mm attributes - it isn't fast (1.8 fps at its highest quality setting), and it isn't an especially good camera for casual or candid shooting due to its size - although I HAVE occasionally used it for this when there was nothing else at hand. It IS a pretty good low-light camera, but low-light cameras as good or better are available for much less money.
     That said, the D3x's strengths are quite dissimilar from traditional 35mm strengths, and make it appealing for landscape photography that would have traditionally been the preserve of medium (and even large) format. The first thing you notice about a D3x image at low ISO is stunning per-pixel detail. Handled carefully (ISO 100 or below, tripod mounted), the amount of detail in a D3x image is even higher than you'd expect from 24.5 mp. Most DSLRs don't look good at 100% on screen, because the blurring from the anti-aliasing filter is apparent at 100% magnification, even if it is not noticeable in a print. Properly focused D3x images look very good at 100%, indicating that the AA filter is either very weak, or alternatively, a new design. Thankfully, the autofocus is very precise, because, in common with all DSLRs, the D3x has a very bright focusing screen that it is not really meant for manual focus. 51 focusing points allow for a great deal of compositional freedom. The best way to focus this camera manually is probably through Live View, which I haven't learned to use yet, having never had it before.
     The second quality noticeable in D3x images is enormous dynamic range - quite simply, this camera offers more range than any other I have used, and by a significant margin. While DxOmark's raw numbers are wrong - the D3x has more like 11+ stops of usable dynamic range than 13 - the differences between cameras captured in their testing seem about right to me. I have extensive experience with the Canon 1Ds mkII, and the D3x really does seem to have between 1.5 and 2 stops of additional DR over the (already pretty good) older Canon. A collection of older DSLRs I've used seem to fit in in roughly the same way - DxO is consistently optimistic by between 1.5 and 2 stops of DR, but their relative rankings are fairly consistent. I mentioned the superb dynamic range in my initial impressions of the D3x, and numerous posters said that 11 stops seemed awfully high - 1500 landscape images later, I'm convinced that 11 stops is, if anything, slightly conservative. In Zone System terms, the D3x goes roughly two zones farther than the original design of the system accommodates - picking up basic tone differentation all the way from Zone 0 to X (its 11 stops of DR), leaving paper white at Zone XI and untextured black at Zone -I. At least 9 of those 11 stops record good detail (as compared to 7 zones with detail in Ansel's original formulation). Medium format digital backs may well be able to match or better this dynamic range, as can Fuji's low resolution SuperCCD, but I have not seen a conventional DSLR that can come close.
    A third characteristic of D3x images is beautiful, smooth tonal transitions. There is none of the tonal harshness (and, in extreme cases, posterization), that plagued many earlier DSLRs. This is aided by absolutely noiseless shadows at low ISOs - something that I have not seen from any other camera I have used. While I have not used medium-format digital systems extensively myself, I have seen quite a bit of MF digital output (I came close to buying into the H3D system before choosing the D3x), and the D3x files really do look like MF files to me, coming strikingly close to the detail, tonality and dynamic range that so impressed me in H3D sample files.
    In overall image quality, the D3x easily exceeds any camera I have used before, although I have not used other "third generation" 24x36 mm DSLRs. There is no comparison between large prints I have made from D3x files and prints from SAMPLE files I have seen from the Canons or the Alpha 900 (the D3x is significantly better than both), but I have no idea how those samples were produced, and I suspect the workflows were not optimal - especially in the case of the Sony, where it is known that Sony's own RAW converter (which was almost certainly used to produce the samples) is far from an optimal choice for the camera. The only large prints of comparable quality to D3x prints I have personally seen come from either medium format digital or large format film.

Comparisons to other cameras I have personal experience with:
Nikon D200 - NOT even close (nor should it be) - I include the D200 only because I have had a great deal of experience with it and know its responses well. In addition to enough more detail that a D3x print at 24x36 inches appears more detailed per square inch than a 12x18 from the D200, there are close to three stops of additional dynamic range in the D3x file. The noiseless character of low-ISO D3x files only adds to the differences! Of course, this is a patently unfair comparison - a consumer DSLR from 2005 against the finest DSLR made in late 2008.

Canon EOS 1DsmkII - a more interesting comparison, because it is the same type of camera, only one generation older. The D3x appears to have approximately double the resolved image detail of the Canon (a 24x36 inch print from the Nikon is equivalently detailed to a 16x24 from the Canon). 150% of the detail is explained directly by the difference in resolution, while I tentatively attribute the rest to the Nikon's superb AA filter - the 1DsmkII has a reputation for strong AA filtration sapping some subject detail, while the D3x has a very weak AA filter that removes very little detail. The image quality difference is even slightly greater than the difference in detail would suggest alone, because the D3x's noiseless shadows and extended dynamic range both contribute to overall image quality. I was never completely satisfied with the 1DsII's shadow quality (it is quite good in midtones and highlights), and the D3x alleviates those concerns.

Hasselblad 501 C/M, scanned with a Nikon SuperCoolscan 9000 - Even Fuji Velvia 100 at 6x6 cm cannot reach the detail of a D3x file at low ISO. While resolution is probably higher, film grain and loss of sharpness from the scanning process interfere with image detail below the detail of a D3x file. Additionally, the dynamic range of Velvia does not even approach what the D3x is capable of, although tonal transitions are equally smooth. More or less the same applies to film I have exposed with a Horseman technical camera at 6x9 cm, although lens resolution was probably more of an issue with the Horseman.

Cameras I have either seen good samples from, or manipulated files from that I did not shoot:

4x5 inch film - A dear friend is a 4x5 landscape photographer, and I do most of her printing with her. Even scanned on a consumer scanner (Epson V700), 4x5 Velvia carries significantly more detail than a D3x file. The tonal smoothness of a 4x5 transparency is simply amazing as well, better than anything a D3x can manage. The one place where the D3x has an edge is in dynamic range - overall image quality of the 4x5 transparency is significantly higher as long as the scene fits within its dynamic range, but the D3x can pull ahead in a contrasty scene.

Hasselblad H3DII/31 - I haven't printed from these files myself, but am judging by large prints on display at PhotoPlus, part of the Hasselblad Masters showcase. I also have not had an opportunity to examine these prints alongside a D3x print (the D3x was not yet released at PhotoPlus). However, the prints in the Hasselblad Masters showcase should have been a reasonable indication of what the H3DII is capable of - it's a mature camera with well-known files, and Hasselblad has many years of experience presenting prints to photographers who know what to look for. Both the tonal gradations and dynamic range are comparable between the H3DII/31 and D3x, and the resolution difference is roughly what the numbers would suggest - the H3DII/31 has a small amount of additional detail (25% or so).


Cameras that should be in this comparison, but aren't, because I either haven't seen samples or don't trust the samples that I HAVE seen:

Canon IDs mk III, 5D mkII - Canon didn't have large samples in a clearly marked location at their PhotoPlus booth, so I cannot be completely sure that any given image wasn't, say, a 50D sample. I also didn't pay as much attention to the Canon samples as I did to the Hasselblad (and Sony) samples.

Sony Alpha 900 - Sony DID have large samples printed from converted RAW files at PhotoPlus, clearly marked as from the Alpha 900, but I didn't trust that they were truly representative of what the camera could do. If the samples were "true", the Alpha 900 stands between the 1Ds mk II and the D3x - resolution closer to the D3x, but with noise reducing image quality significantly, and with dynamic range in between the two and a tonal scale more like the older Canon. Even this is a very impressive performance for a $3000 camera - it clearly beat a camera that sold for $8000 the year before. However, there are several caveats about the PhotoPlus samples that make me think the Sony could do even better under other circumstances. First, it was VERY new at PhotoPlus - it was actually not yet for sale in the US at the time (it arrived about a week later) - who knows what stage of final development the cameras that made those samples were in? Second, the Alpha cogniscenti both here and on DPReview don't especially like Sony's RAW converter, which is almost certainly what was used to convert those files - Capture One and RAW Developer are both supposed to do a much better job. I wouldn't call the camera and workflow that made the suspicious samples ready for 24x36 - although it could certainly do 16x24 with no problem. This was clearly a late beta or early production camera that nobody knew how to handle yet, up against fairly well optimized workflows on the cameras I've worked with myself.

Reading other people's thoughts on the Alpha 900 (perhaps the most interesting competitor, due to the common pedigree of the sensor - while not the same, they're clearly related) , I'm inclined to believe that the image quality difference is there in RAW files, but it's not huge - some cleaner shadows at low ISOs increasing dynamic range, plus high ISO performance. Some reviewers who've used both, including Michael, believe that there isn't a difference (most are Alpha shooters who have only quickly picked up a D3x), while others believe that there is a significant difference (most are Nikon shooters who have less experience with the Alpha). My best guess is that, given two different photographers - one for whom the D3x is his or her chosen tool, the other one the Alpha 900, and the same subject matter, the D3x will have a slight, but visible advantage on certain subjects in an optimized workflow. However, with ANY camera at this level, every photographer will get their best results with their own camera - if I picked up an Alpha 900, I couldn't get what I get out of the D3x after months of experimentation, and if an Alpha shooter picked up a D3x, they wouldn't get results to match their Alpha.

The more significant difference between the D3x and the Alpha 900 is in the type of body they use. The Alpha 900 body is a near-direct copy of the $1100 Alpha 700, which is itself partially a hodgepodge of pieces from various Konica-Minolta film SLRs, ranging from the very well respected advanced-amateur Maxxum/Dynax 7 to several $300 consumer SLRs. I couldn't get comfortable with an Alpha 900 quickly (I've never used one for an extended period, and I'm sure I'd do better with time), and there are a couple of places where its sealing worries me - it uses several slide switches which are just about impossible to seal, and its card and battery doors don't lock. In addition, the Alpha 900 uses consumer-level AF (although folks who use it say that the center AF point is truly superb) and metering systems, while the D3x uses the best in the business. A large part of the $5000 difference in price between the two cameras goes to the bulletproof body of the D3x, with its 50 year professional heritage going back to the original Nikon F (although that doesn't explain why you can buy that SAME body with a lower-resolution sensor in it, plus an Alpha 900, for the price of a D3x).


Medium format above the H3DII/31 - I didn't see any samples clearly marked as coming from a high-resolution back. I would believe that the 50 MP backs are just about as good per pixel as the D3x without a problem (leading to twice the overall image detail). If that is true, twice the performance is available for 4-5 times the price, and it'll take a whopping print/printer to see it to its best advantage. Of course, the D3x itself is extracting a little bit of extra performance (plus a very different type of body) for a big premium over a 5D mkII or an Alpha 900.

One element of the D3x that should excite even Nikon shooters who don't like the price or the size is the way Nikon tends to reuse a sensor. Unlike Canon, who use innumerable sensor and resolution variants across their line, Nikon tends to go for "variations on a theme" - they made nothing but 6mp and under cameras for years, then jumped straight to 12mp with the D2x. Later, they explored the 12mp theme very thoroughly before jumping straight to 24mp. If this is the theme on which Nikon will build three or four years worth of variations, it is a darned good basis to start from.


The D3x provides the detail and tonality necessary to make high detail landscape prints up to 24x36 inches, and it provides it in a rugged and feature-laden body that contains every option any photographer might want. It is a joy to work with in the field, including in harsh conditions, and is easier to carry than what we used to lug around to make the same type of images (which doesn't make it light). Is it worth $8000? To me it is - I will probably never buy another high-end camera for its features - what more could I need? Finally, we have reached a point where resolution, dynamic range and tonal scale are  so good that, even for large, highly detailed prints, we can focus on the art instead of the equipment.

Does a "better" camera exist? Yes - almost certainly - but it costs five times as much. I wouldn't want to be any of the MF digital manufacturers right now, with the D3x and the other 20+ MP DSLRs coming so close to their entry-level 31mp products. The high end of medium format is safe, because the laws of physics will reduce per-pixel image quality if anyone tries to cram much over 30mp on a 24x36mm sensor, but the market of photographers who need to print OVER 24x36 for close viewing is tiny, and the market for $30,000+ bodies is smaller still. Just as the MF manufacturers were making some progress on the pricing of their lower-end models, 24x36mm SLRs infringe on their image quality.


What about the lower-cost cameras with most of the D3x's image quality? The 5D mkII seems to have some durability worries, plus it has greatly inferior AF, metering and other body features. It's a viable option in the studio, but, especially after the Antarctica report, I wouldn't want to take it into the field the way I use my D3x. The 1Ds mk III is comparable to the D3x in body quality, but also fairly close in cost, especially with its recent price increase - Nikon leads in image quality right now, but the next revision of the 1Ds will probably regain the lead for Canon. Choose between Nikon and Canon's top bodies based on the rest of the system - do you prefer Nikon's wides, or Canon's teles? Do you prefer the speed, durability and low-light capability of the D700 as a second tier body, or the resolution of the 5D mk II? The most interesting competitor is the Alpha 900 - it's better made than the 5D mk II, its image quality is superb, and it has several unique features like body image stabilization, all for a very reasonable price. The question in my mind with the Alpha is Sony - both Canon and Nikon have been selling and supporting top-level professional cameras for many years, and both have developed full systems to support those cameras. Minolta has been in and out of the pro market since the original Maxxum 9000, but never demonstrated a sustained commitment - they would develop an interesting camera like the 9000, the 9xi or the Maxxum 9, but they would never have a full line of pro lenses, flashes or other accessories. Will Sony take a sustained interest in the top end of the market, and in the needs of the photographers who buy these cameras? Sony has a VERY interesting body, and hiring Carl Zeiss to make their top-grade lenses was a great move - perhaps they are going to take a sustained interest in this market? If so, I wouldn't be surprised to see an "Alpha 1000" - a camera like the 900, but with top-end AF, metering and more complete seals (no slide switches).
 
Right now, the D3x is simply the best DSLR available, taking image quality, durability and features into account. It's certainly not the camera for everyone, as much because of weight and features as price. For those of us who work rugged and print big, it offers a unique combination of attributes that go a long way towards justifying its price. It can easily replace medium-format film, and compete with low-end MF digital, at a cost lower than either (assuming a large number of images per year). It is supported by the 50-year Nikon system, with an incredible range of lenses and accessories for any task, plus other bodies that compliment the D3x for other purposes.

                                               
Any of the three 20+ mp systems are so good that they can handle just about any photographic task (especially when combined with other bodies with different strengths in the same line - pick one, learn it well and make the best images you can with it. I've preferred Nikon handling and glass for many years, so I was excited when they released a SLR with this image quality, but both Canon and Sony make viable alternatives for photographers who prefer those systems.
                                                           -Dan



Logged

jing q

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 596
    • we are super
Nikon D3x review from a landscape shooter's perspective
« Reply #1 on: February 21, 2009, 01:19:20 am »

I cannot believe you wrote that long post to defend a camera.
There must be something better you'd like to do with your time.
I cannot believe you compared it with Velvia either. Velvia is known for its intense colours. I always use a milder film like 100G if its details I'm looking for mainly.
Also, a Nikon coolscan 9000 is NOT a good scanner for medium format film
Even an Imacon draws out much more detail from a piece of film, what more drum scanners? And you can find good deals on Imacons.

Also, not everything is about resolution.
Alot of medium format shooters don't even print very big, it's more about the look and feel of the picture and the PROCESS of shooting (if you've looked down into a RZ67 viewfinder or a Hassy viewfinder you would know exactly what I mean)
This is all the more important with landscape photos. It's the ability to visualize on a 2-dimensional screen how you see the world that is a big plus.
There's a reason why people shoot 4x5 also. It creates a different state of mind when you see an image 4x5 inches in front of you while composing.

also how can you compare a 1DsMkII with a D3x??I myself own a MkII and have used a MkIII and the MkIII blows the MkII out of the water. It's not a good comparison.

btw, a "better" camera does not cost 5 times as much. You can pick up a 22mp back for less than 10k nowadays. that's about the same price as a D3x. The MkIII files I've shot are nowhere near the quality of the 22mp back files. Tonal transitions, a certain meatiness to the files, etc etc.

while I appreciate your love for Nikon, I find it hard to stomach this drum thumping thread. You sure Nikon is not paying you?

Quote from: Dan Wells
In overall image quality, the D3x easily exceeds any camera I have used before, although I have not used other "third generation" 24x36 mm DSLRs. There is no comparison between large prints I have made from D3x files and prints from SAMPLE files I have seen from the Canons or the Alpha 900 (the D3x is significantly better than both), but I have no idea how those samples were produced, and I suspect the workflows were not optimal - especially in the case of the Sony, where it is known that Sony's own RAW converter (which was almost certainly used to produce the samples) is far from an optimal choice for the camera.
Logged

dalethorn

  • Guest
Nikon D3x review from a landscape shooter's perspective
« Reply #2 on: February 21, 2009, 02:16:38 am »

I appreciate the post, since you have a lot of information from actual use that's missing in most reviews. There's a lot of arguability about price -vs- value, as the articles on the main site attest. Still, I don't think direct comparisons of price are always useful - I would prefer to know 1) can I afford it? 2) is it a good enough performer compared to the competition to justify spending more? 3) does it have lasting value? 4) is it a real professional design? 5) is it durable?

It seems in its weakest area the D3X is best in its size and price (more or less) range, and in its other stronger areas, wins hands down.

Still, the Fuhrer video was funny as all heck.
Logged

dwdallam

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2044
    • http://www.dwdallam.com
Nikon D3x review from a landscape shooter's perspective
« Reply #3 on: February 21, 2009, 02:32:35 am »

Nikon's latest release is the king it seems. I've played with a Dx2 (the 12MP pro camera) unless I have the numbers wrong, but you all know what model I'm talking about, and I very much liked the way it felt in my hand and it's overall balance better than either my 5D or my 1DS3.

Canon may match or exceed the D3X in the coming months, and we'll see if Nikon can respond, or if they are running on empty now with their latest effort. But one thing is a fact--from what I've experienced, and that isn't much--the Nikon bodies feel better and more solid in my hand. They are not more solid, but they just feel like it. In other words, their ergonomics are better, for my hand.

I hope Sony doesn't stop offering Nikon the option to build their sensors based on economics though, or Nikon will be in trouble. The next camera I buy I'll be very careful about it. If Canon lenses don't equal Nikon's across the board, and I'm ready to upgrade, I may sell my Canon kit and go purely Nikon, and enjoy the superior ergonomic feel I get from the Nikon. The next time I'm ready, however, Nikon will have to shine as much as it is now against Canon. If they are equal, I'll most likely stay Canon, eat the ergonomic superiority of the Nikon, and save myself a few thousand in lenses and equipment related to Canon Bodies. The Canon f2.8 70-200L is still a powerful reason to stay with Canon, if one uses that lens a lot.

On the other hand, when I got into this game, Canon was the clear winner in resolution, image quality, and everything else, except ergonomics and fps. It seems like all things come around and go around. I'm really not too worried about it at this point.

Dan was, and for all the reasons he stated. He needs every last ounce of general overall image quality for the work he does. He got that finally with the D3X, and so I understand his post. Is the Nikon D3X really the best FF camera? Probably. I haven't read anything yet that says it's not that's a valid test. Anyone else?
Logged

jing q

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 596
    • we are super
Nikon D3x review from a landscape shooter's perspective
« Reply #4 on: February 21, 2009, 02:50:56 am »

I'm just slightly amused at the need to defend a camera for certain uses
if you're going to spend $8k on a camera for landscapes which are locked down and low ISOs you're better off with a full frame medium format digital back.
No AA filter, richer tonalities
Logged

dwdallam

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2044
    • http://www.dwdallam.com
Nikon D3x review from a landscape shooter's perspective
« Reply #5 on: February 21, 2009, 03:35:39 am »

I don't think he's "defending" anything. He's just telling us his experience as it relates to his needs.
Logged

NigelC

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 583
Nikon D3x review from a landscape shooter's perspective
« Reply #6 on: February 21, 2009, 06:24:20 am »

Quote from: jing q
I'm just slightly amused at the need to defend a camera for certain uses
if you're going to spend $8k on a camera for landscapes which are locked down and low ISOs you're better off with a full frame medium format digital back.
No AA filter, richer tonalities

The D3X may be judged not a cost effective solution by some relative to other 21Mp offerings, but looks good value compared to a 22-31Mp MFDB solutions if it come very close on IQ. (say £10,000 for body and 3 lenses, compared to c£20,000 fro H3D31 plus 3 lenses)
Logged

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1044
Nikon D3x review from a landscape shooter's perspective
« Reply #7 on: February 21, 2009, 09:31:26 am »

From what I've seen,the D3x beats any (used would be the only option) MFDB in the same price range. It loses (slightly) to the H3DII for twice the money, and probably loses more decisively to options that cost four or five times as much. Also remember that the D3x (and the EOS 1Ds series) offers durability that is in a totally different class from MFDB solutions - MFDBs have to be babied in the field, while the D3x is a brick. Many older-generation MFDBs, plus the current low-end Mamiya/Leaf, actually have fans, which strikes me as the opposite of the D3x's dust sealing (the Phase backs ARE very well sealed, as are the current-generation Hasselblads, but the bodies that go with them are not).

In reference to not using the prosumer CCD scanners, what other option is there that doesn't cost MUCH more than a D3x and take up a great deal more time as well? Even if there is detail in 6x6 (and especially 6x9) transparencies that exceeds what a D3x would get on the same scene, what good is that detail if it takes hours per image and a $50,000 scanner to extract? That becomes an unlimited-budget option like the P65+ and the H3DII/60. Also, the Super Coolscan is scanning film grain when it scans medium format at maximum resolution, so a drum scanner can't really get much more total detail - I have no doubt that it WOULD reach deeper into the shadows, and I'd be interested to see D3x shadows (which well exceed anything a moderately priced scanner can do) against a drum scan.

                                                        -Dan
Logged

sc_john

  • Guest
Nikon D3x review from a landscape shooter's perspective
« Reply #8 on: February 21, 2009, 12:02:15 pm »

Dan,

You believe that D3x produces an excellent print to 24"x36". Do you have opinion at what print size the difference between D3 and D3x would become visible... resolution, dynamic range, etc.? Thanks.

John
Logged

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1044
Nikon D3x review from a landscape shooter's perspective
« Reply #9 on: February 21, 2009, 12:21:24 pm »

I haven't used a D3 myself, but I'd guess the difference would become obvious somewhere between 12x18 and 16x24, unless circumstances strongly favored one camera (a scene that used the D3x's extra half to one stop of DR, or needed the D3's speed). I can say that the D3 breaks away from consumer SLRs between 8x12 and 12x18 - I can easily tell it from $1000 SLRs at that size, even in scenes where its wide DR and tonal response don't give it away. There are certain circumstances where the difference is visible even at 1024x768 on screen - if the D3x has held a highlight or retained detail in a shadow that another camera wouldn't have (the same applies to the SuperCCD Fujis, which also have very wide DR - in the right scene, the difference is visible at any size). If the scene dynamic range doesn't give it away, a good consumer DSLR will stay in the race through 8x12, but have visibly less detail by 12x18. The 16.7 mp Canon 1Ds mkII will clearly lose detail compared to the D3x by 16x24 (but will hold up at 12x18 if DR and tonal scale aren't major factors), and my best estimate on the D3/D700 is somewhere in between consumer 10-12 mp SLRs and the 1Ds mk II.

                                       -Dan


Logged

sc_john

  • Guest
Nikon D3x review from a landscape shooter's perspective
« Reply #10 on: February 21, 2009, 01:19:22 pm »

Thanks... well reasoned write-up and follow-up.

John
Logged

douglasf13

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 547
Nikon D3x review from a landscape shooter's perspective
« Reply #11 on: February 21, 2009, 02:57:13 pm »

As someone who owns cameras from various entities across a couple of different formats, I'm not really entrenched in any of the companies, but I'm currently using Sony more than anything else, so I'd figure I'd offer a couple of rebuttals to your A900 mentions, and I'll say now that I think the D3x is an outstanding camera, so I'm not taking anything away from it.  

  I believe you are absolutely correct about the A900 and RAW converter choice.  Unfortunately, IDC and ACR are the worst two converters out there for the A900, and it seems they are also the two most popular choices for tests.  Just about anything else, from Bibble, to C1, to RPP, to Raw Therapee, provides much better output at all ISOs, and this is especially the case at ~ISO 800+.   FWIW, the two reviews that I've seen for the D3x and A900 that give actual resolution numbers give a slight edge to the A900, although it's hardly worth mentioning, and lens choice is the more important issue in that regard.

  After talking to dual D3x/A900 users, like the Borgs, I think it should be mentioned that, while the D3x IQ may be the most desirable overall, there are reasons why shooters such as Iliah keep both cameras (outside of the ZA lenses.)  According to him, while the D3x is the better camera for high contrast landscape scenes with it's lower shadow noise, the A900 actually performs better in flatter landscape scenes because of it's color separation.  He also believes that the A900 is a bit better for dark haired and dark skinned portraits.  Now, I think we should all take these observations with a grain of salt, but I think that the greater point is that it is really difficult to determine a "best" camera, because these camera's have nuances that may or may not be noticed, unless they are used side by side. That being said, it appears that the D3x has slightly better noise in the shadows and slightly better higher ISO, and that is important to many.  Interestingly, if you're shooting a low contrast scene with the A900, ISO 400-500 actually gives slightly better shadow noise performance than the lower ISOs, which is an interesting trait of the Sony's ADC.  Also, assuming you're using a good converter, shooting at ISO 400 and underexposing seems to provide slightly better results than shooting at ISO 3200, which is counterintuitive to many, and usually not worth the hassle to me.

  As far as build quality, I think your characterization of the Sony being a Minolta parts bin camera is a bit outrageous.  The A900 is a much better built camera than the A700 ( I own both,) with a magnesium back, magnesium chassis and better sealing.  None of the "creaks" that I've noticed in my A700 exist in the A900, and I'm not worried about the slide switch issue that I've seen you mention many times.  As Michael R. alluded to in his Antarctica write up, I wouldn't hesitate to use the A900 (or D700, for the matter) right next to a D3x user in any conditions.  In terms of reliability, I'd much definitely choose 2 A900's over one D3x.

  Finally, in regards to AF, you hit the nail right on the head.  There isn't a camera out there with a better center AF point than the A900, but it can't compete with the Dxxx or Dx series in regards to outside AF points and tracking.  That is a big advantage to many.

  Ultimately, I'd agree that the D3x appears to be the best 35mm DSLR on the planet overall, but if two shooters, one with an A900, and one with a D3x, went on a trip together, I think each camera would exhibit IQ advantages over the other, depending on the scene, and it seems lens selection would provide the more striking differences, which is a good place for cameras to be.  Cheers.  -d
« Last Edit: February 21, 2009, 02:58:09 pm by douglasf13 »
Logged

Professional

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 309
Nikon D3x review from a landscape shooter's perspective
« Reply #12 on: February 21, 2009, 03:14:00 pm »

I hope this thread will not making me to regret that i went with Hasselblad H3DII and not Nikon D3x.
Logged

douglasf13

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 547
Nikon D3x review from a landscape shooter's perspective
« Reply #13 on: February 21, 2009, 03:15:47 pm »

Quote from: Professional
I hope this thread will not making me to regret that i went with Hasselblad H3DII and not Nikon D3x.

 I wouldn't worry.  That H3Dii is a sweet setup.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Nikon D3x review from a landscape shooter's perspective
« Reply #14 on: February 21, 2009, 03:17:19 pm »

So...

Pixels are free but slide scans cost 50 box each?

Erik

Quote from: KLaban
I don't pay for drum scans, my clients do.

A good drum scan will extract the finest detail whereas the prosumer CCD scans will bloat that fine detail and ultimately loose it.
Logged
Erik Kaffehr
 

jing q

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 596
    • we are super
Nikon D3x review from a landscape shooter's perspective
« Reply #15 on: February 21, 2009, 03:58:08 pm »

Quote from: Dan Wells
From what I've seen,the D3x beats any (used would be the only option) MFDB in the same price range. It loses (slightly) to the H3DII for twice the money, and probably loses more decisively to options that cost four or five times as much. Also remember that the D3x (and the EOS 1Ds series) offers durability that is in a totally different class from MFDB solutions - MFDBs have to be babied in the field, while the D3x is a brick. Many older-generation MFDBs, plus the current low-end Mamiya/Leaf, actually have fans, which strikes me as the opposite of the D3x's dust sealing (the Phase backs ARE very well sealed, as are the current-generation Hasselblads, but the bodies that go with them are not).

In reference to not using the prosumer CCD scanners, what other option is there that doesn't cost MUCH more than a D3x and take up a great deal more time as well? Even if there is detail in 6x6 (and especially 6x9) transparencies that exceeds what a D3x would get on the same scene, what good is that detail if it takes hours per image and a $50,000 scanner to extract? That becomes an unlimited-budget option like the P65+ and the H3DII/60. Also, the Super Coolscan is scanning film grain when it scans medium format at maximum resolution, so a drum scanner can't really get much more total detail - I have no doubt that it WOULD reach deeper into the shadows, and I'd be interested to see D3x shadows (which well exceed anything a moderately priced scanner can do) against a drum scan.

                                                        -Dan

hi dan, the thing is, the 31 mp backs are cheaper than the 22 mp backs because in terms of sheer quality the 22mp backs with their fat pixels are still something really gorgeous to look at. I shot with a H3D 22mp back on a loaner and I found the quality stunning. the 31 mp backs have higher pixel density which is not necessarily the best thing. in other words if a 31 mp H3d beats your D3X, even more so a 22mp 645 frame back.

I think you need to try one out for yourself, just to check out the files for yourself.

like I said, a 22mp back is about the same price as a D3X right now if you look on ebay. or even cheaper. Yes you have to get the camera though. I'm using a AFd + lenses , which is only a few hundred bucks each.

I got a imacon flexphoto on ebay for less than $2000. it will definitely extract much more detail than a coolscan 9000. the 9000 chokes up the shadows very fast, contrast is very high. I never liked it. Sharpness isn't the best either due to the design of the holder.
I do my scans, they take 5 mins tops.
When I shoot landscapes I only end up with 1 or 2 photos from a set which I keep. I don't see the point of shooting film for stuff that requires multiple shots, but with landscapes you usually don't need to have alot of frames to choose from.

hope you don't take offense at what I've been saying and I really wouldn't mind a D3x myself, but given the choice between using my MFDB and a D3x for a landscape on a tripod, I would take the MFDB anyday.

Logged

David Sutton

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1345
    • David Sutton Photography
Nikon D3x review from a landscape shooter's perspective
« Reply #16 on: February 21, 2009, 04:28:15 pm »

Many thanks Dan for your thoughtful article. Such an amazing time to be photographing. David
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Nikon D3x review from a landscape shooter's perspective
« Reply #17 on: February 22, 2009, 12:33:15 am »

Hi,

Have you actually compared the 22 mp D back to a Nikon D3x? The Imacon seems to be interesting, I don't shoot film any longer , but I used to have problems extracting all information from slides using my DSMP (Minolta Dimage Scan Multi Pro).

Best regards
Erik

Quote from: jing q
hi dan, the thing is, the 31 mp backs are cheaper than the 22 mp backs because in terms of sheer quality the 22mp backs with their fat pixels are still something really gorgeous to look at. I shot with a H3D 22mp back on a loaner and I found the quality stunning. the 31 mp backs have higher pixel density which is not necessarily the best thing. in other words if a 31 mp H3d beats your D3X, even more so a 22mp 645 frame back.

I think you need to try one out for yourself, just to check out the files for yourself.

like I said, a 22mp back is about the same price as a D3X right now if you look on ebay. or even cheaper. Yes you have to get the camera though. I'm using a AFd + lenses , which is only a few hundred bucks each.

I got a imacon flexphoto on ebay for less than $2000. it will definitely extract much more detail than a coolscan 9000. the 9000 chokes up the shadows very fast, contrast is very high. I never liked it. Sharpness isn't the best either due to the design of the holder.
I do my scans, they take 5 mins tops.
When I shoot landscapes I only end up with 1 or 2 photos from a set which I keep. I don't see the point of shooting film for stuff that requires multiple shots, but with landscapes you usually don't need to have alot of frames to choose from.

hope you don't take offense at what I've been saying and I really wouldn't mind a D3x myself, but given the choice between using my MFDB and a D3x for a landscape on a tripod, I would take the MFDB anyday.
Logged
Erik Kaffehr
 

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1044
Nikon D3x review from a landscape shooter's perspective
« Reply #18 on: February 22, 2009, 08:45:46 am »

I agree with Douglas that the differences between the D3x and A900 in image quality appear to be small enough that a photographer who uses either one (and knows it well) will get better images with their own camera (with known workflow and quirks) than with the other camera - I don't think I could quickly pick up an A900 and do it justice (in a couple of months, I'd be getting really nice stuff, but not the first time out). I wish Sony (and, frankly, Nikon) would just bundle C1 with their camera for RAW conversion - Phase is open to bundling deals with everybody from Leica to SanDisk to Microsoft! ACR doesn't tend to do a great job on most Nikon files any more than it does on Sony files - because it historically hasn't been great, I'm not excited enough about it to bother fooling with the beta. When Aperture (which has been superb with many Nikons) is out for the D3x, I will give that a try. I'm lucky enough to have owned C1 Pro for years, always keeping it around for certain conversion tasks, even when using something more convenient as my primary converter. 4.6 is much easier to use than 3.x, so I'm likely to stick with C1 as my primary converter, at least until and unless something much better comes along.  Even the newest C1 is still not as good an integrated image management system as Aperture or Lightroom, although the cataloging aspects of the program are certainly MUCH improved from C1 3.x. One thing that C1 (at least C1 Pro) does VERY well is produce multiple conversion types from the same image at the same time - I have it set to make a jpeg I can e-mail in addition every time I create a print-ready file.
       I'm intrigued by the observations on Sony's build quality (I hope you can tell from my review that I respect the A900 and welcome a third player in the market - I hope Sony stays around, because three high-end makers will drive innovation) - I thought that the 900 was known to be basically a 700 body except for the pentaprism (and, of course, the sensor), but it seems like it's not, despite the nearly identical control layout.

                                                                               -Dan
Logged

douglasf13

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 547
Nikon D3x review from a landscape shooter's perspective
« Reply #19 on: February 22, 2009, 11:40:34 am »

It sure would be nice if C1 was bundled with Sony.  I'm curious about the difference in the way ACR treats D3x vs. A900. I saw a comparison of ACR with the A900 and 5dii, and while ACR wasn't the absolute best for the Canon, I was amazed at how much better it did with the 5dii files.

 As far as A900 build goes, it is a bit surprising to me as well, because I was expecting an A700 type build. Don't get me wrong, the A700 build is competetive in it's class, but when you work with a camera for a while, little niggles like body and card door creaks become noticeable (I've heard about these in the 5dii already.) The A900 uses some new labyrinth seals in the doors, and it just feels more solid, albeit it's no D3x.  

  The only issue Ive had with either the A700 and A900 in lots of use is that, once or twice a year, a little compressed air or contact spray is needed to clean the control dials, because they get a little wonky.
Logged
Pages: [1] 2 3   Go Up