« What, Me Argue? | Main | Keep On Learnin' »

Wednesday, 17 December 2008

Sony vs. Nikon vs. Canon

...That is, the A900 vs. the D3/D700 vs. the 5D Mark II.

I admit to feeling considerably unhappy about having to summarize my own use of these three cameras "in public." I'm sure to disappoint almost everyone. "Brand partisans" being what they are these days, I'm probably going to take it on the chin, too. I'm even a bit miffed at myself, actually. Of all the tiresome endless squabbles in photography's little corner of the internet, the hoary old "Canon vs. Nikon" debate probably ranks as #1, and I kinda promised myself I was never going to get into it. How exactly did I, then?

Anyway, I've decided that rather than carefully annotate all of my tests and trials, which wouldn't begin to convince anybody anyway, I'll just come out with it and say what I think. But you've got to promise me you'll read this next bit and remember it as you go along, please: I'm not the arbiter. I don't get the final say. I'm just one photographer...one more, out of many. And I haven't bought, and am not going to buy, any of these cameras, so anything I say is not brand partisanship on my part, or self-congratulatory pride of ownership, or anxiety that I might have made the wrong choice, or insecurity that somebody else might own something better than mine, or fear that having a "less than the best" camera might reflect on my bad photography, or any of the other motives that fuel all the passionate nattering.

Sigh. The above is not going to protect me, I know. Had to try. For the record, I don't care to name a "best" because I'm not all about status, and I could care less whatever anyone personally chooses to use. It's the picture that counts.

And now, I gotta do what I gotta do.

The ultimate in image quality: Belongs to the Sony A900. That's assuming you're someone who knows how to use it and is willing to use it as it's intended to be used—it's not perfect for every style of work. But between its staggering resolution and very good dynamic range, its willing response to the Exposure and Recovery sliders, and its more "photographic" image quality (more on this later) and lack of digital artifacts—and despite its less-than-accurate color—it's the IQ emperor for now, among these four (i.e., the three under discussion and the D3). Downside: ya gotta need (or crave) that resolution. Otherwise it's just an also-ran. But if you do, you're just not going to get more impressive image files out of a 35mm-styled and -sized camera body.

The most recommendable: This prize goes to the Nikon D700. Given its sensible size compared to the D3, robust build, fast autofocus, overall responsiveness, superior ergonomics, unmatched high-ISO performance, and perfectly sensible file size, it's going to be the most bang for the buck for more photographers than either of the others. The Nikon is flat-out a better camera than the Canon, a point exemplified by its clearly superior autofocus performance (the 5D Mark II, when asked to autofocus, is sort of like "Huh? Oh, right," whereas the Nikon is a snappy "Yessir!") Its image quality is really pretty, and its files are superbly printable, and my feeling is that it would help more photographers take better pictures in more situations than either of the other two.

 The best compromise: And here's where we come to the Canon 5D Mark II. No, the Canon does not have the A900's ungodly resolving power; but it comes reasonably close. And no, sorry, no matter what you've heard here, there, and everywhere, the Canon does not match the Nikon's high-ISO performance. Don't just look at the noise; look at the pictures. (I say that having made myself bleary-eyed poring over the most inane "test shots"—something I also once promised myself I'd do as little of as possible.) But it's pretty close. And here's the thing: it has much more resolution than the Nikon, and much better high-ISO capability than the Sony. So its win over the Nikon where resolution is concerned is bigger than the margin by which it loses to the Sony in the same department, and its win over the Sony in high-ISO performance is much more decisive than the margin by which it loses to the Nikon on that score. As they say in auto racing, when you're really good but not quite the best, what they call you is second—but strong seconds in two categories could be a good thing.

My personal choice: ...Would be to take the plunge, be a mensch, and make the call between the D700 and the A900, with their opposing but unsurpassed strengths. For many others, I'm sure, the Canon's "strong compromise" will be the way to go. (And bear in mind I've completely ignored the Canon's video capability, which many shoppers will not do.)

And to put myself on the hot seat rather than take the fence-sitter's way out ("all three cameras are fine devices with formidable imaging capabilities, and all three deserve your strong consideration"), I had to ask myself the question two ways. First, if I were spending my own scarce, hard-earned money? That would probably tilt me towards the Nikon—better support, more mindshare, greater corporate stability (at least in the camera division—once burned, and all that), much better market penetration, probably more robust resale value—plus I'm seduced by the D700's lovely B&W abilities. But if somebody put all three cameras with their respective 35mm ƒ/2 lenses on a platter and invited me to take one, I don't think I could resist the siren call of the beautiful Sony. It's a bit of a head/heart split.

Markandmegsmall
In very low light at some friends' Christmas party, the 5D Mark II and 35/1.4 acquitted themselves splendidly, giving the D700 an Alydar-like run for the money.

Artifacts
Here we go.

I have to admit I have some mild reservations about the 5D Mark II's image quality. It's very good, no question: Canon has hit the "what consumers want" targets on the nose: More Megapixels! Less Noise! (Great Taste, Less Filling). But there's at least a partial price to pay for all that tasty goodness. It shows up in the form of what I'd group under the heading "artifacts."

It's been eight or 10 years since "purple fringing" (also called "CA," not entirely accurately) first heaved into our collective consciousness as a peculiarly digital anomaly, and since then, other artifacts have been dealt with in their turn. I don't see much in the way of purple fringing at all from the 5D Mark II, but there's what Carl Weese calls "blue replacement," by which narrow objects imaged against a brighter background change from their own color into a darkish pastel hue. You see it most often in twigs and telephone lines. The 5D Mark II isn't particularly bad, but it shows up a lot more than it does from the D700. And its susceptibility to blue replacement makes it a candidate for a lovely lens aberration that I'd never actually seen before in a picture I've taken myself—longitudinal chromatic aberration (LoCA), which shifts objects in front of the plane of best focus to magenta and those in back of the plane to green.

Picture_16
Canon EF 35mm ƒ/1.4 LoCA on the 5D Mark II

Then there's the quality of the Canon's noise. It's a bit tilted towards the chroma type, and it has a weird, blotchy character. (I don't read the forums—is the consensus that Canon is applying noise reduction to the raw file?) Note that the following is a small section of a much larger picture.

Picture_19
The 5D Mark II at ISO 5000 and, after you click on this image to open it, 100%

Here's a section at 300% (again, that's after you open it) to show the character of the noise:

Picture_20

Of course, this is pure pixel-peepery...you're never going to see the character of the noise even in imposingly large prints. I'm less sanguine, however, about the Canon's color performance at high ISOs. It seems to resist correction more than I'm used to, at least in ACR.

Then there's an entirely new artifact that, as far as I know, is unique to the 5D Mark II: sometimes you'll see black fringing next to blown highlights, but only on the right-hand side. Weird. This shows up fairly often in the shooting I've been doing at night that includes Christmas lights.

Finesnowsmall

You can see it in the detail of this shot, from near the left edge. Look at the highlights at the top and bottom of this detail:

Picture_37_2

This doesn't bother me at all—you'd never see it in prints and you probably wouldn't notice it if it were visible—but hey, I'm just a reporter, I gotta report what I see.

I think we can be reasonably certain that Canon will fix this in a future firmware update.

Finally, there's highlight clipping. This is probably my most serious reservation about the Canon because it really does affect the look of pictures. Here I have to go back to the Sony A900, which is particularly good in this respect. This is a detail (maybe 30%) of a larger image, and this is somewhat behind the plane of focus (the subject was a girl lying on the bed*), so don't expect it to be sharp—just look at the tonality (luminance values):

Picture_25
Sony A900 detail

It's just a screen shot from ACR, no corrections applied. Now look at the same section with some judicious use of the "Exposure" slider and the "Recovery" slider slammed all the way to the right (you can see that it's picked up a bit more blue, for one thing):

Picture_26

Nothing startling. But the above is a pictorially convincing result to me. The transitions are soft, natural, the rendition of the bright light convincing. You expect the area near a lamp to be somewhat blown out...there are three areas here that remain blown out, but they all transition naturally, and you even see just enough detail in the stem of the lamp below the shade (which is white, even) to indicate visually what it is. Tonally, I can't say I'd demand more than the above from Tri-X, which is high praise for a digital camera. This is one of the real strengths of the A900 and one of a number of reasons why I give it my personal palm for IQ.

The 5D Mark II's transitions to blown highlights, by comparison, seem to be more like a throwback to the era when 6-MP DSLRs were the SOTA**. They tend to be harsh and abrupt, and less fixable.

Canon conclusion
The 5D Mark II is a brilliant camera in many ways. It has more, and better, of most of the good things we like. But my holistic overall impression is that its images just look more digital, in some fundamental way, rather than just "photographic."

To test that impression, I compared it with what is becoming my "Old Trusty," the Pentax K20D, a currently $810 camera with a really good APS-C sized sensor (and which I also don't own, N.B.) The 5D Mark II is better than the 14.6-megapixel K20D. But it's not that much better. If the 5D Mark II lags behind the D700 by not-quite-a-stop in high-ISO noise, then the K20D lags behind the 5D Mark II also by not-quite-a-stop—certainly it looks at least as good at ISO 1250 as the Canon looks at ISO 2500 (and yes, I ran the tests). And the Canon does indeed have more resolution—but not that much more resolution. A paper size, maybe? Maybe a tad less?

In my very personal estimation (remember what you promised in the third paragraph of this post!), the A900 and D700 seem like game-changers. They both strike me as enough better than the prevailing standard that they create new paradigms—in very different ways. The Canon 5D Mark II can simply be looked at in two ways. From the positive side, it gives you the lion's share of what the other two cameras do best. And that's good. And from the negative side, it sort of seems like just another digital SLR, only better (and please don't quote that line out of context). Which way will you see it? That I can't say. All three cameras are fine devices with formidable imaging capabilities, and all three deserve your strong consideration....

____________________

Mike

*A canine girl. Don't get excited.

**state-of-the-art

Send this post to a friend

Only 7 shopping days till...oh, never mind.
Amazon U.S. link
Amazon U.K. link
Amazon Germany link
Amazon Canada link
B&H Photo link
Adorama Camera link

Featured Comment by Jeff Baker: "After reading every reviewer's opinions, it seems a leap of faith must still be made, by each shopper individually, and then the real fun begins. There are subtle differences that 100 reviewers might not notice that may make or break your world. The camera is the front end of the workflow, and a lot can be done downstream to enhance, fix, or adjust whatever issues these cameras might have.

"I have an original 5D, and the more I read, the less inclined I am to upgrade to anything. I'm pretty much convinced my 5D has some magic qualities I've not seen elsewhere, and it produces prints up to at least 2x3 ft. without problem. Not sure what else I need. Yep, it doesn't like to focus in very low light, and maybe has other issues, but there's lots of other places to spend money."

Featured Comment by sfwrtr: "Interesting article. I did note that you used ACR to do the Canon 5D Mark II conversion. I have discovered massive defects in the ACR 5.2 release in regards to this camera. The bundled Canon DPP application, though unwieldy, does a better conversion from RAW with significantly fewer artifacts and improved tonality. I don't know about the fringing, but CA compensation is part of the DPP process and it might handle the green and purple problem (dunno). You might want to check it out. I must say I am annoyed at Adobe for releasing my beloved ACR converter too soon, and so is the Canon rep I've spoken to. By the way, I've posted an example of the difference as an animated GIF.

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00df351e888f88340105367a6b11970b

Listed below are links to weblogs that reference Sony vs. Nikon vs. Canon:

Comments

Well done, Mike. Thanks for taking the time to share your impressions. There is much information to ponder.

Thanks Mike. Your camera reviews are the only ones that I find both informative and damn hell hilarious.

Well done Mike!
Thank you for taking the bull by the horns and sharing your (enviable) experience with these three cameras with the us. I know someone will roast you for having an opinion. Personally though, I like hearing what you have to say whether it be photographic, jazz, or just life. The world would be a boring place and the arts would not exist if we all thought the same thing.
So, keep it up!
Thanks,
john

Aaiyyeee. Could I be the first to comment?I guess I really SHOULD be out shoveling.
I'm a Canon guy and I'm already disappointed that the 5DII isn't holding up to the D700. I have friends with the Nikon and others that have used the 5D to some dismay.
..but to comment on the "game changing" bit...as a crabby Canon user I'm keeping in mind that they will kick back and feel confident that the "industry" (whatever that means but I mean photojournalism) is going to latch onto the new, consumer-accessible ability to also shoot VIIIDeeeeee-ooooooHHH and Canon will at some point down the road claim to have been there first...(not counting RED).
Thanks for kickass reviews nonetheless Mike. I personally don't mind at all when a respected reviewer says "The Canon just isn't as good as the Nikon...or the Sony" I'm overly invested in Canon glass....no going back now.

good non-biased comparison, good insight.

ryan

www.ryanhollowayphotography.com

Since you didn't reccomend Nikon as the hands down winner and totally slam the other two cameras I quit reading the post early on!I'm sure most of your readers will do the same or else they are wrong as well! Seriously, isn't it amazing to have so many really wonderful choices? I'm only glad that I lost all of our disposable money, and a lot of our non-disposable money!, in the current financial fiasco or else I would have to explain to my wife why Santa brought me a SonyA900 when I allready have a D300 and a D700!Keep up the good, objective, and most importantly geared to people who actually take photographs with their cameras reviews!

I appreciate your attempt at being unbiased in your assessments of the big three (not automakers;-)I have been a long time admirer of Freeman Patterson, a great Canadian photographer and teacher, who, if am correct, used a Minolta, and noone claims his photos are not good because he did not use a Nikon or Canon, or some German make. So if all three are that good I would choose the cheapest camera. In my case, I would choose Nikon because I already have Nikon manual lenses which I am loathe to give up. I am convinced the autofocus zoom lenses, particularly long zooms, are vibration prone.

What I'm interested in is your impressions that some images are "more photographic" and/or "less digital". Like the old thing where people thought they could recognize a medium-format print instantly by the smoothness, or something, I think it would benefit everybody if we could characterize these differences more accurately. I'm not very good at seeing them, either; either I'm not critical enough, or something (my eyes are very unlikely to be bad enough to physically prevent me from seeing such things). If we can characterize them better, we can have an idea of what the actual differences are.

The only bit of the medium format differences I could consistently see was different grain presentation for same-size prints from different-size formats.

(As a D700 owner, I don't feel you've told me I did the wrong thing :-). The AF is really nice, even when you're using the high ISO.)

There are no bad choices here and each is better than most of the photographers who will buy them.

Thanks for the essay Mike.
As a D700 owner, I can agree with most of what you say. As a photographer, I'd just like to add it's a great time to be taking pictures.
Modern digital SLRs have revolutionized the way we shoot. Who would have thought that 6400 would ever be an acceptible sensitivity?
I was speaking to dear old Dad the other day -who was a professional photographer his entire life- about this, and I could practically see the gears turning in his head, as he imagined the possibilities. He was reminiscing about having to make do with ISOs (or ASA, as he still calls it) of 25, or even 8.
I guess my point is, is that because the market is so competitive today, camera makers have to come up with a great product -and for the most part, they do.
Sometimes it seems that a lot of these comparisons are really just hair splitting contests, with the ultimate winner being us, the shooters.

Remember Mike, when this is all over, that you are loved.

Hi Mike,
Thanks for making some great points re the Sony and Nikon. I would be interested to hear what you think of their black and white capabilities, especially as compared to film.

Three amazing tools. In the hands of an amazing artist, any of these are capable of helping create wonderful art.

Its an embarrassment of riches, isn't it?

You're right regarding the 5D2 files, they look too digital. So I have finally decided to buy myself a...

...Zeiss Ikon, Biogon 35/2 and loads of film!

Thanks
Lars

Mike, A great assessment of the three most desirable cameras out there. I'll admit to being a Nikon shooter but I held the Sony in my hands recently and it was the first digital camera that gave me "tactile camera lust" in a big way. The finder was absolutely gorgeous. Now if I could just have it all. Alas, I have a 13 year old who wants to go to college some day........

Thanks for the nice read!

Kirk Tuck

You are one brave sonofabitch, Mike. I wouldn't have the stones to post that...but thanks for doing it.

A number of thoughts:
This thing called noise, as applied to digital
photography, we didn't have that issue blatently pitched, did we?

And the other thing; being able to expose a
sensor or light recording device ie film in almost black out conditions and
then being able to reveal an image without the assistance of additional artificial illumination, what is the idea? To do what may have been impossible prior and because such an activity is now within the realm of possibility?

Suspect we are all wanting more and more. Greed?
I don't know.

I do make one expensive note; the Sony, Nikon and Canon are all EXPENSIVE devices, more so when a piece of glass is added to the mixture.

Yes, you reviewed them however in this mad race to produce an improved image capture device, most of us are aware the devices reviewed will be considered disposable when the next latest and greatest appears.

Is our rapidly declining consumer society really prepared for such a vast tumble?

And given the so-called stupid decline in our economy, maybe it is time for all of us
to gather our wagons in a circle, and wait out the battle for our way of life.

I'm about to buy the 5DMk2, so i'm a bit concerned about your comments regarding its noise and 'digital-like' files. I knew about the black dots, but don't believe they'll ever affect me. The noise and digitalness, though - well, that surprises me. I had felt that my old 5D imaged in a more 'organic' manner than any Nikon. I'm sorta hoping that your assessment may be based more on Nikon familiarity and not on an empirical truth....

Anyway, i will no/cannot argue with any of your comments. But, i will offer that the reason i cannot choose in favor of Nikon, despite my preference for their body design, is that Canon's prime lenses are all just so much better. At least, in the range i'm likely to use (28mm-85mm). And, frankly, i remain shocked that Nikon has not chosen to address this. Nikon still doesn't have a fast 28 or 35mm lens. Their 50mm lens has the worst bokeh of any 50mm lens i've ever seen, and their new G isn't much of a step forward. The 85mm/1.4 is nice, but still doesn't approach the 85L.

If i could have a D700 body, with Canon glass, and the 5D2's resolution and video, i'd just stop. Right there. But, i'm going to have to deal with the 5D2's 'shortcomings.' I never had a problem with the old 5D's AF. I don't shoot sports/action, though. And, in fact, when i compare Canon to Nikon AF in shops, i tend to prefer the Canon's. Go figure. The Canons just seem more immediate and responsive, and the Nikons seem to hesitate initially, before then speedily locking on. I'm sure that's a familiarity issue, too.

You qualify the A900's ultimate IQ rating by saying, "That's assuming you're someone who knows how to use it and is willing to use it as it's intended to be used." So, then, what is it's intended use?

Mike
Thank you for your well reasoned and valued opinion - the reason I think the blog is so popular and respected. It's a rather sad reflection that you felt in neccessary to pre-contextualise your post - I hope your chin survives...

Dominic

"So, then, what is it's intended use?"

Michael,
You would use the A900 in situations where you didn't need extremely high ISOs and did need lots of resolution, for large prints, for example. It would not be the camera I'd want to shoot basketball games in poorly-lit high school gymnasiums for half-page repro, for instance. It would definitely be the camera I would want to make 20x24 fine art prints for framing and sale. And so forth.

Mike J.

"I think it would benefit everybody if we could characterize these differences more accurately."

David,
That's what I tried to do in the section called "Artifacts."

Mike J.

What some smart poster at DPReview figured out:

The excessive purple fringing on the 5DM2 results from reflections, ghosting, double imaging, from the filter stack that sits in front of the sensor.

He figures whatever they did to change that stack from that which is on the 1DSM3 in order to make it transmit more light (for better high ISO) is also causing this fringing effect.

There's almost no way firmware can fix this one.

* * *

The black dot effect of the 5DM2 is a separate one from above, and it's unknown whether this can be addressed in firmware.

* * *

Bottom line is to get the high ISO performance at such a resolution, Canon pushed a couple of limits here and probably pushed over the limit by a touch. So you have to ask yourself, knowing this, am I OK with this?

I don't think I am. :(

I have Canon and Nikon Systems and I think they are both great. I have used them for a while and in the real world have difficulty picking a favorite or deciding which is "best." I tend to favor Canon overall. I have found that with all cameras I have tested, there is a great difference in performance dependant on the conditions and it is impossible to test all conditions. I do think that video capabilities will play a much bigger role in the future, and I bought the 5DII largely because it was an inexpensive way to be more creative (lenses) in video. It turns out it is a great still camera too. Also, most purchasers of high end SLR's will be influenced by the lenses they have or that are available. I think Canon has the edge on lenses, but many would argue that point as well. I personally believe that in the next 5 to 10 years we will have low noise at very high ISO (50,000), but right now, I am very happy with around ISO 2000 with all these cameras. And resolution is good enough right now, but will improve as technology progresses and/or sensor size gets bigger. As an audiophile, I know there is a limit to what I can hear, and with cameras, there are also limits that will eventually make further improvements moot. In the last few years, we have been presented with the best imaging tools in history and competition had something to do with that. This is a great time to be a photographer.

Well done, Mike! I hope we can look forward to hearing more from you about the lenses you used on these cameras.

So, (knowing nothing about Sony lenses), Mike, It sounds like the Sony A900 would be the current best pick for landscape shooting, below medium format?

Like everyone else, I follow your personal impressions of these cameras with interest. Your knowledge base and skills as a photographer are so much more than mine! But when all is said and done, I keep scrolling back through your blog entries to take another look at the "Unscared of the dark" shot you took with the Nikon D700...wow! I'm a Canon shooter with a few pieces of decent Canon glass, but if not for that I think this one picture would persuade me differently. BTW, how did that shot print out? Thanks for the time and thought you put into these "reviews".

Pffft!
The international industrial conglomerate that makes the electronic consumer product that I purchased is better than the international industrial conglomerate that made your electronic consumer product!

Ok, each one is a wonferful camera, I just wolud like to keep the A900 in my hands for few second.....and hear his "click".....Absolutely Fantastic!

It's always a delight to read some clear-headed thinking on such a potentially nebulous subject. One of the reasons I haunt this site rather than others (cough-DPReview-cough) is the consistently thoughtful approach to cameras and image quality here. Rather than ruminations over sub-pixel-level phenomena, we get this kind of sensible discussion about what really matters when the ink hits the paper.

It's also probably fair to point out how absurdly good all 3 of these cameras are. Even using the best possible photographic technique, grain and tonality issues still place a limit on the largest print you can make from a 35 mm film image—at least one that looks good to my eye. Yet all 3 of these cameras let you print a carefully exposed and processed image to crazy sizes—and have it look perfect. The low-light capability of the D700 is simply mind-boggling to someone who grew up on Kodachrome, where the "really fast version" was ISO 200 and had golf-ball sized grain. And the resolution attainable from 20+ megapixel sensors (with appropriate glass) is magical. The only excuse for a bad picture is, sadly, the photographer.

Time to take a few more pictures and practice so my camera won't embarrass me again.

Considering the investment I have in Canon lenses and accessories, articles like this are an intellectual curiosity and not much else. If I want to go to a full frame camera, it will be the 5DmkII. I'm not prepared to sell what I currently have and by a new system based on what's out there now. If there is truly a camera that advances the state of the art by a thousand fold, I would consider it, but none of these cameras do.

I've made my bed and I'm comfortable in it.

"The lamp sample shots show that the Sony had overall less exposure, so it's not really directly comparable, IMO. Also, not knowing if they were shot in JPEG or RAW, how they were processed leaves more questions as to whether we are looking at real differences in D.R., or just differences in processing and exposure."

This is always a problem when writing reviews. It's the difference between opinion and evidence. What I want to do is give my opinion, or impression, and then illustrate with an example to make what I'm saying more clear (if only because a written article without illustrations is dull). What you want to do is take the example and make it into evidence. And as evidence it's never enough, because it's only one particular sample.

To demonstrate my impression--to give evidence--would require a great deal of work, in fact more work than I have done. I would have to make direct comparisons, carefully bracket the exposures, use different converters to remove that variable, etc. And then people would say, "yes, but maybe it's all down to ---------- or --------- and you never tested that." That's why I prefer to simply shoot with the cameras and just give my impressions instead. But it's a mistake to treat an illustration as evidence, and it's why photo writers prefer that you pay attention to what they say over what the illustrations show you. At least with our words, we can better clarify how and in what way our opinions might be wrong.

Mike J.

It's really Sony vs. Canon here. Nikon is not a contender, because I don't want a 12MP camera for 3 grand, and I don't want a 24MP camera for 8 grand.

I shoot 95% wide angle. Canon's wide angle lenses are good but not great in the corners. When the Zeiss ZA 16-35/2.8 lens arrives next year for the Sony, I'll have to take a serious look at the A900 before deciding between that camera and the 5D Mark II.

Somebody up the line spoke about the "photographic" less Digital style.
I remembered a comment somewhere, yes, here it is:

http://www.swiatobrazu.pl/_and_they_compared_to_one_another_hasselblad_h3_mamiya_afdiii_sony_alpha_900_821.html,5

Also good read with a different perspective.

Thanks; Mike, good comparison :-)

One thing you left out of your assessment is the phrase "For the moment." The A900's resolution will almost surely be eclipsed by Canon within the year, and has already been matched by Nikon. That means that a new buyer, planning to use the camera "as it's meant to be used," would do well to think about the implications of buying into the system, because Sony's resolution advantage is almost sure to be fleeting. Of course, the Sony's other huge advantage is cost, which may be critical to many people.

On the other hand...the Sony will make exceptional pictures for many years, and below some certain print size -- 24 inches, perhaps? -- it'll be a long time before anything surpasses it, if it's "used as it's meant to be used."

By the way, I really like the head shot of the young woman in what looks like low tungsten light; I even like the blotchiness. Pretty sensual shot. Like a Dutch painting where the varnish has gone yellow with the years.

JC

Yes, I should have said more explicitly that you had made some attempts to characterize the issues, and that I applauded that while encouraging more of it. I didn't intend it to sound like I was criticizing you for not doing it.

Thanks for the review - it was very informative, and more or less what I expected: Sony having the most resolution, and Nikon, with its relatively low resolution, able to squeak out best high-ISO performance. I tend to take the stance of the 5D II being "best of both worlds" since it's so close in both departments, but I already have an arsenal of Canon glass so there's probably a cognitive bias lurking somewhere in there... ;-)

I'm curious about two things: first, like others asked, what you mean by "digital" vs. "photographic" pictures? On the surface this sounds like a very logical, reasonable assertion, that might be responded to similar to the light fall off question, but surely there's more to it than that.

Last: have you tried down-rez'ing the 5D Mark II image files to the size of a D700, and then comparing noise? The 5D M2's noise gets significantly less noticeable when it's downsized to the equivalent of the D700, and I suspect is at least the equal to the D700 in that case (can't perform the test personally because I don't have a D700 handy!). There's actually an SRAW option that shoots directly in ~10 MP mode, too, which has less (noticeable) noise.

Mike, just this Tuesday I saw several A4 printed images from Nikon D700. I looked at them very closely (like in 2 feet from my eyes, or may be even less). IMHO, for printing up to (inclusive) A3 one does not need more than what Nikon D700 provides. Since many people who are amateurs like me rarely print bigger than A3, I have to ask myself (and you too in a manner of speaking) do we, the amateurs, really need more than 12-15 MP? I think that we don't. Unless heavy cropping is involved, I don't think that these MPs are indeed justified. It seems however that right now we're square in the midst of megapixel wars. To that end I applaud Nikon's decision to come up with D700. Since we (virtually) know each other for quite some time, I should say here that I am sorely tempted to go Nikon. D700 seems to be exactly the photographic tool I need.

Meanwhile, tomorrow I shall be giving in my freshly bought FA 24-90 for AF repairs.

Reading your blog is very joyful to my eyes.

"That's what I tried to do in the section called "Artifacts.""

But you only show artifacts from one camera in isolation. Where is the lamp highlight shot for the 5D2 and/or Nikons? Where is the comparable fringing or noise shot for the A900 and Nikons? Could fringing simply be the lens (e.g. use the same Nikon MF lens for both the Nikon and the Canon.

As a long time Nikon shooter I always felt Canon had superior AF. I'm just wondering if in your AF test were you using the 35 f/2 or 1.4L on the Canon?
Thanks for telling it like it is Mike!

Hi, thanks for the report. The lamp sample shots show that the Sony had overall less exposure, so it's not really directly comparable, IMO. Also, not knowing if they were shot in JPEG or RAW, how they were processed leaves more questions as to whether we are looking at real differences in D.R., or just differences in processing and exposure.

Thanks for this review. Since we all get more or less locked into one system or another what with lenses, lighting systems, etc, I ponder if I'll regret spending for an upgrade rather than which of the three excellent cameras to choose. My opinion is I wouldn't regret upgrading to any of these 3.

All 3 of the systems now go into the 20+ megapixel range. I'd be thrilled if the camera makers now turned their attention to competing on high ISO noise at their current generous resolutions.

Finally, I realize this is sacrilege, but I've come to like the digital look.

I'm only echoing most of everyone else's statements in that I am highly grateful that you've taken the time to test and summarize your findings to the rest of us.

I think nature of the shooting you do matters a lot more as we get into these high end cameras. Having recently moved to the D700 from a Canon, I am amazed at how much better a camera it is for shooting things that move in lower light - not darkness, but decent indoor light. Plus all the necessary controls for tweaking shots on the fly fall right to hand (finger), for very fast shooting and adjustments such as exposure comp and shooting mode (P A S M). OTHO, on a tripod shooting still lifes and landscapes and architecture, the Sony and Canon are going to produce much nicer files for really big prints - if you make really big prints. For myself, I still shoot 4x5 if I want to use a tripod and make really big sharp prints. The D700 is prefect for me because it is a camera for people and difficult light when I do not want to use a tripod. If I quit shooting 4x5 and was only going to shoot digital, I would probably go for the Canon or Sony.

Great, fair reviews. Nice work. All of these cameras look to be top flight. I wish I could have told you not to use ACR with the A900 when you had it, because just about any other RAW converter works much better with the camera's noise. For some reason, Adobe just can't seem to get it right with the A700/900.

The best part of this (though i enjoyed it all) is that it makes me happy about being able to get a Pentax K20D next year to go with my Pentax 35 F2.

Time to check the couch for spare change again.

Cheers!

Great review! I shoot with Canons, used to shoot with Nikons and I've owned a couple of Sony...TVs.

I've got pictures hanging on the walls at home and at work made with 4x5 Wista, Canon 1n and 30D, Leica M6, Pentax 645, Mamiya C330, Nikon F2 and Kodak disposable. I love all these pictures and I consider the camera used in making them to be the least important part of the whole process. But it's still fun to read about which car won the race this month.

So my question, as I contemplate purchasing a 5DMKII, is whether it might come closer to having "the ungodly resolving power of the A900" with some new Zeiss ZE mount glass in front of it.

At PhotoPlus this year I took a look at the Zeiss 85mm for EF mount, and it was certainly impressive through the viewfinder--sharp with really smooth out of focus rendering. Unfortunately, they only had a mockup of the 50/1.4 in ZE mount at the show.

I have always purchased Canon SLRs (after being an owner of Olympus point-and-shoots, which depsite having lousy sensors, STILL meter as well or better than any other brands' cameras) but I work with a staff of Nikon users.

To say that Nikon's new SLRs are amazing sells them short. From the D40 through to the D3, they beat their Canon counterparts in sharpness, noise control, ease of use and just about every other measure. Canon still makes great cameras. Nikons are just superior and I'm intrigued that Sony has come so far so fast with their cameras.

It's a great time to be a shooter.

Good summary Mike. I have used all these and the D3X as well, currently under review. My experience with the 5D MkII is not as bad as yours, and despite some very poor Canon glass issues (unsharpness to one side or the other) with two lenses I have found it hard to make the A900 match the visual acuity of the 5D MkII. The technical reason seems to lie in the A-to-D conversion curve, which is pushing shadows down then lifting three-quarter tones and midtones (I think this also results in the sharpening artefact seen as the 'black dot').

The A900 has, as you say, some qualities which induce loyalty - mainly the viewfinder. It's better than the D3X in brightness and size, and also in dioptre setting tolerance.

I like the rubber waterproof flash shoe cover on the D3X. It makes a statement.

David

The comments to this entry are closed.