See also the additional comments further below added 18 Dec.
18 Dec: Update for those who misread and misinterpret my posts, and particularly to those who have no knowledge of the voluminous past writing on the Sony cameras and their appeal and the praise I have offered (always in context of why).
- I like the cold, but when I do not explicitly say I also like it dry hot such as when cycling, people assume that I *only* like the cold. Reactions to my posts sometimes fall into this category even though no such thing has been stated or even implied.
- While this is the age of the internet (knee jerk reactions and context dropping), I write for those willing to expend some degree of intellectual effort in understanding the context I am writing in, that is, for Miller’s Law to apply: “To understand what another person is saying, you must assume that it is true and try to imagine what it could be true of”. Many of my readers do so, but in my lifelong experience, few people apply this principle or even approach it. Indeed, I have to remind myself, even being conscious of it, but it is extremely practical and useful skill that greatly expands the bounds of learning.
- I like the Sony A7R III. Just because I point out badly done things (mainly because they stick out like sore thumbs) does not mean I dislike a camera, let alone condemn it. Indeed I would recommend it first to most people most of the time! But in context, it is my view that having to pay $3300 to replace a Sony A7R II is a lot of money, that much of what is changed in the A7R III could and should have been done in a firmware update for the A7R II a year ago, and that the gains of the A7R III (for me) so far appear to be modest (at best) and thus it is a weak argument in terms of value in my usage context.
- I will have to buy the Sony A7R III for business reasons. It is a very complex camera with many positives, and also negatives. Since the negatives pop out immediately, they get mentioned first. But why they are there at all speaks to poor design review, and thus they deserve attention, first. My reviews are staged, not lump-sum.
- People forget that I am not a sports photographer or street photographer and never have claimed to be, and that any kind of photographer does no one a service by pretending to judge a camera outside his/her field of expertise (ditto for general “camera of the year” context-dropping awards). I scrupulously stick to areas that I feel I can add value.
- Mixing the appropriateness of a camera system with one’s ego is a fundamental error—spending big bucks is a commitment that inherently biases nearly everyone. Some people can set this aside, others cannot. Automobile sales personnel know this, which is why they will try to sell you on any make or model to start with—it jumps a psychological hurdle that makes it very difficult to reverse course and lubricates the path forward to more options, better model, etc. It relates to why any criticism of a camera is taken by some to be a personal affront. I say only to such people: recognize your lack of objectivity as a disservice to yourself.
- I would not pound nails with a screwdriver. For landscape, for architecture, for whatever, there are camera(s) that are best for each of those jobs. For landscape, dynamic range and robustness of pixel quality are driving factors for me, along with ease of operability in a variety of conditions, including cold down to 15°F or so. For someone in Florida, cold is not a factor at all! Context-dropping is an invalid basis for disagreement—that is, throwing away premises and context of usage. It’s like arguing whether apples are better than tuna-fish.
- When the key goal is “small and light”, I can’t see any possible rationality in being irked by anyone (like me) stating that I greatly prefer the Fujifilm GFX or Nikon D850 for landscape. Because my premise is “best image quality and operability and size/weight balance”, whereas (I have said this many times in past writing), size/weight may be the driving factor for some and for me in some conditions.
- If I were walking the streets of NYC, I would not be carrying my D850 or a GFX; I’d be carrying a Sony A7R III. Similarly, if I were climbing a peak by myself or with people I want to capture, I’d carry the Sony A7R III.
On 17 December (the day after writing the original post below), I hiked most of the day with the Fujifilm GFX + 45mm f/2.8 along with the Sony A7R III and 4 of the best lenses available for it. I shot zero images with the Sony. Why? Because with the GFX, I know I will be rewarded no matter how difficult the contrast and that I can aggressively control contrast right down into deep blacks. Not so with the Sony. Call it a gut reaction to all the Sony and Fujifilm images (thousands) I have shot in the past years. Peak image quality was my priority.
In Michael E’s case below, his context (which I know well) totally justifies his choice of the D850 and makes the Sony an also-ran exercise in frustration. It is just not the best tool for what he does. But the Nikon D850 would be a poor choice for someone else in some other context.
... Original post below ...
I meant what I said and I said what I meant in my trailing comment in yesterday’s post:
Having shot the Nikon D850 and Fujifilm GFX heavily in the past 6 weeks, I’ll say this plainly: the Sony A7R III trails far behind in satisfaction in every area for the way I shoot.
Nor is image quality at the same level in things that count (noise in dark areas for example as well as the character of the noise). Like the A7R II, the Sony A7R III has its small and light uses, Eye AF, etc—really winning features in some cases for sure—but for landscape shooting it leaves me utterly without desire to shoot it. I dread having to spend the money to buy it and would be perfectly content saving that money and sticking with the A7R II.
Michael Erlewine writes:
Along with Sony’s abysmal support for dubiously implemented pixel shift, my shooting needs are much akin (though different subject matter) to those of Michael Erlewine, who writes:
I did some further tests today using one of the esoteric Nikon lenses, the Printing Nikkor 150mm, version one. I tested both the D850 and the A7R3 to see what differences there are, and found nothing to write home about with the A7R3. I did NOT test the pixel-shift because I am kind of disappointed in it. It does not seem worth the effort. I’m sure it can be shaped-up with practice, but the question I ask myself: is it worth it? Right now, the answer is “No.”
I keep falling into thinking that the A7R3 is just a toy, although I know it is not. But after several years of testing all of these new cameras out, I am impressed about how good and professional the Nikon D850 is. I would be better off buying two of them, one for a backup, but I have a D810 which is still working fine.
So, I believe I am going to send the A7R3 back, with as many trimmings as I can, and sell off the rest, which is a bunch. This includes the Voigtlander 65mm macro which is a great lens. I am done testing Sony cameras for a while. You would think they could hire someone like you to build a really useful camera, but they don’t.
So, it is time for me to just spend some time taking photos with the Nikon D850 and be happy with that. I am tired of also-ran cameras for a while.
I read your notes on the A7R3 and agree. It just makes me feel tired, like Sisyphus pushing uphill. I am just getting too old and weary to enjoy Sony’s learning curve. I think I may just return the whole thing and turn into a person happy with my Nikon D850, which means NOT trying every last camera that purports to be an improvement. I am very happy with the D850 and hate to waste the time fiddling with the Sony A7R3. I kind of resent them putting me through this. LOL.
DIGLLOYD: Sony A7 series are hardly “also ran” but what Michael means is for his needs. And ditto for my shooting needs. Nor is any of the foregoing a criticism in any way of what works best for someone else.
That said, there is huge room for improvement on hardware and software with Sony mirrorless, which is mostly a conglomeration of (very cool) tech—more a computer than a camera in too many ways.
Comments following the original post above:
Donald L writes:
Needless to say as an owner of the A7R II and III I have been closely following your impression of the new Sony offering. I completely agree that the Sony conversion software is something that should never have been released . It is an embarrassment to a company wishing to be respected by the professional photography community. I believe it would have been better for them to rely solely on Capture One; which I personally find a most credible offering for many imaging systems.
The last time I owned a Nikon camera was film. I did try using a very early digital model but found the noise and resolution to unacceptable. Before purchasing the A7R I thought about going back to Nikon with the D850 but even in view of the superior ergonomics and quality of the images, I could not imagine going back to a mirror box after focusing my Zeiss lenses directly off a sensor. Personally, I believe that the D850 is at, or near, the end of the line for DSLR and look forward to see what Nikon offers in the future with mirrorless. Hope all your earlier heath issues resolved and keep the good work coming.
DIGLLOYD: I focus all of my landscape images off the sensor using the 2.4 megapixel rear LCD and the Zacuto Z-Finder 3X loupe. I realized that’s an issue for other types of shooting, but see Focusing Zeiss DSLR Lenses For Peak Performance parts 1 and 2 where I show in particular in one video how Live View can be used for even street shooting.
The mirror box (aside from lens design constraints) is simply not a factor for me for what I like to shoot—I use it for initial framing to find exactly the right shooting location (in 3D terms), then put the camera in that precise location with a tripod (well, barring those 30° tricky slopes I might have to compromise sometimes). After that, it is all Live View.
Thom Hogan writes:
D850 or Sony A7RIII? The D850. Nikon nailed this camera as much as they ever nail a camera. Sony is still missing critical elements that make a camera fully usable in the field (With apologies to Don Ho: "Tiny buttons, on the body, make me feel unhappy, makes me feel shoddy..."). From the standpoint of ergonomics alone, the D850 just is in a different class than the Sony. Sure, smaller and lighter is a Sony win, but it comes at a big handling price.
Image quality? Still a Nikon win, though a smaller one. Sony still has some strange minor anomalies in their sensor. Fixed pattern noise is one if you get too aggressive with ISO or pushing shadows. Compressed raw still reduces bit count and produces highlight edge artifacts. I don't know what Nikon did differently with their sensor, but I was surprised to find that, yes, pixel level output is improved from the D810. The D850's raw pixels are among the easiest to work with I've encountered, and the low level detail is an improvement from the D850 despite the modest pixel count increase.
As for Pixel Shift (A7RIII) versus Focus Stack (D850) ....
.... Hopefully you didn't buy either camera for these latest and greatest technologies, but to just take great images. Both can do that, but the Nikon D850 is just better.
DIGLLOYD: a reader pointed out this post to me late on Dec 18.
Thom Hogan is well worth reading, and he does different things than I do: more wildlife and action-oriented stuff. Read Thom’s entire post. But he should not conflate pixel shift (tiny improvements with Sony, distinctly betterwith Pentax K-1) with the massive gains of focus stacking. They are different things entirely, and unless an image is resolved to the sensor resolution, pixel shift has next to nothing to offer. And the only way to resolve to the sensor is to use focus stacking (stacking pixel-shifted images can be done, but is well nigh impossible outdoors most of the time, due to subject movement or just a faint puff of wind on the tripod).
I rank the Nikon D850 as top of the heap for the 35mm format and by a significant margin, though as one reader suggested to me: 'no one will be able to tell the difference'. I agree in this sense: I could claim the same with my iPhone panoramas and it would hold true. 'No one' meaning the 99.9% who don’t even look at images at Retina grade on a Retina display and who think images 2000 pixels across are very large. Since I post my work at full resolution 7360 to 8600 or so pixels across, I have a bit different perspective. And 10K displays are now in the HDMI 2.1 specification (59 megapixels). If the new Mac Pro doesn’t support 8K I’d be very surprised, but 10K might be there also. Context matters, and yesterday’s 16-megapixel APS-C images will be tiny letterboxed historical anachronisms, like my 0.9 megapixel digital images from 1999.
In some circumstances the Sony A7R II/III files break down to brittle unworkable junk, whereas the Nikon D850 files hold their integrity—this is why people want to argue about it—they don’t 'get' what happens with real field shots but prefer studio shots where there are no challenges. Even with my silver van shot in bright sunlight, the Sony A7R III shows unbelievably bad noise patterns if sharpened even a little too much—and on the brightly-lit silver areas which ought to be ultra-clean with beautiful nearly noise-free imagery—not so—the ugly orange peel crap that debuted with the original A7R is still there. The A7R II/III are just not in the same league as the D850 when it comes to real world performance—the hell with the GIGO lab tests, SOMA for the masses. And the Fujifilm GFX is a cut above the D850 as one would expect with 50% larger pixels together with Fujifilm’s outstandingly clean raw file quality.