Sony A7R III sensor review: The Nikon D850 meets its mirrorless match

100
DxOMark Sensor
Sony A7R III showing sensor

The first cameras in Sony’s Alpha 7-series of mirrorless full-frame cameras were announced in October 2013 and are credited with bringing compact system cameras (CSCs) to the attention of professional photographers. Since then, they’ve proven extremely popular and helped Sony win a huge chunk of the interchangeable-lens camera market.

The Sony A7R line is the high-resolution offering, and the 42.4Mp A7R III replaces the A7R II, which itself was the replacement for the original (36Mp) A7R.

While the A7R III and A7R II have a backlit sensor with the same pixel count, the Mark III’s sensor is augmented with a front-end LSI that gives it greater processing power and a two-fold increase in the data readout speed. This, combined with the Bionz X processing engine, enables a maximum continuous shooting rate of 10fps with continuous autofocusing and metering.

Other specification highlights include a 5-axis stabilization system that Sony claims gives a 5.5 EV extension to hand-held shutter speeds, and a Pixel Shift Multi Shooting mode that creates a single composite image from four images to provide greater detail. (We made all our measurements using the standard shooting mode.)

Key specifications:

  • 42.4Mp full-frame (35.9x24mm) Exmor R CMOS sensor
  • Front-end LSI and Bionz X processing engine
  • ISO 100–32,000 standard range for stills, expandable to 102,400
  • Pixel Shift Multi Shooting for greater detail
  • 3.69-million-dot OLED viewfinder
  • 399 on-chip phase-detection AF points and 425 contrast-detection points
  • Sony EF lens mount

Overall image sensor performance

Sony A7R III Scores
The Sony A7R III achieves an overall DxOMark sensor score of 100 points, the same as the Nikon D850. This makes it the highest-scoring mirrorless system camera to date, beating the A7R II by two points. Naturally, this means that the sub-scores are also very good, and the sensor can indeed capture wide dynamic range with good color, but it’s the Sports (Low-Light ISO) score that particularly impresses at ISO 3523. This is the fourth-highest value we’ve ever recorded — beaten only by the sensors in the medium-format Hasselblad X1D-50C and Pentax 645Z, and by the 12Mp full-frame Sony A7S II. The size and pixel count of these sensors means that the photoreceptors are in principle significantly bigger than those in the 42Mp A7R III, which gives them an advantage for light gathering and noise control.

Image quality compared

As it’s a full-frame interchangeable-lens camera with 42.4 million effective pixels, the Sony A7R III is a direct competitor to the 45.7Mp Nikon D850. It’s also going to be of huge interest to Sony A7R II users, so we’ll take a look at how the three cameras compare.

Sony A7R III scores comparison

Click here to open our interactive DxOMark comparison tool

As you can see, the Nikon D850, Sony A7R III, and A7R II sensors occupy the third, fourth, and fifth slots in our scoring hierarchy. They are beaten only by the medium-format sensors in the Hasselblad X1D-50C and the Pentax 645Z, which take the first and second slots, respectively.

Although the D850’s sensor has the same overall score as the A7R III’s, it’s ranked higher because of its better Portrait (Color Depth) and Landscape (Dynamic Range) scores (we’ll look at this in more detail in the next section). In normal shooting conditions, the D850’s 0.4-bit higher Portrait (Color Depth) score is unlikely to be noticed, but it might make a slight difference in image processing. The 0.1 higher dynamic range is also so similar to the new Sony’s that in practical terms, they are the same.

While A7R III’s Sports (Low-Light ISO) score of 3523 is excellent, it’s actually only 0.4 EV higher than that achieved by the Nikon D850. So while the A7R III handles low-light situations better and produces images with less noise, the difference isn’t huge.

The most significant improvement of the A7R III’s sensor over the A7R II’s is its dynamic range (Landscape) score — 0.8 EV better, indicating that the new camera can capture a broader range of tones in a single image. That should make an appreciable difference in landscape photos.

In-depth comparisons

In this section, we’ll take a closer look at how the Sony A7R III compares with its predecessor and the Nikon D850.

Portrait (Color Depth)

At its lowest sensitivity setting (ISO 32), the Nikon D850 sensor has a Color Sensitivity of 26.4 bits, a value that puts it on par with medium-format cameras. At ISO 50, its lowest sensitivity available setting, the Sony A7R III sensor has a Color Sensitivity of 26 bits, which is an excellent score and a close match to the D850’s at ISO 100. Above this value, the two cameras produce very similar results for the rest of their sensitivity (ISO) range.

Although the difference is unlikely to be visible in most instances, it’s interesting to see that the Sony A7R II actually beats both its successor and the D850 for Color Sensitivity for much of its sensitivity range. However, at around ISO 1600, the color sensitivity of the D850 and A7R III drops to around 20 bits. As the ISO is pushed above this value, color sensitivity continues to drop, and it will become increasingly difficult to reduce the visibility of noise without reducing color saturation when processing raw files. The A7R II sensor copes a little better, not crossing the 20-bit threshold for about another 0.7 EV.

Our tonal range results show that with the exception of the lowest ISO values, the Sony cameras record a broader range of tones. The ability to record tonal variation drops as sensitivity (ISO) rises, and the Sony camera sensors are about 0.5 EV better in this respect than the Nikon’s.

Nevertheless, landscape, portrait, still-life, and commercial photographers are likely to use as low a sensitivity setting as they can, so if they are able to drop to ISO 32 with the Nikon D850, they will be able to capture a broader range of tones. If they have to use a higher sensitivity value, the advantage is lost and in some respects, the A7R III sensor does a better job.

Landscape (Dynamic Range)

Again, the Nikon D850 benefits from its lower minimum sensitivity value that enables it to capture images with very high dynamic range. At the first few sensitivity settings above that value, however, it captures a very similar range to that of the Sony A7R III before the Sony camera takes the lead at around ISO 640. This lead of about 0.3 EV continues for the higher values, growing to around 0.7 EV by ISO 51,200.

Further good news for low-light photographers is that the A7R III’s dynamic range falls below 10 EV only at around ISO 6,400.

At high ISO values, the A7R II matches the Mark III for dynamic range, but the newer sensor shows its advantage at the lower values, from ISO 50 to 3,200 — the most commonly-used settings.

Sports (Low-Light ISO)

As with the Portrait (Color Depth) and Landscape (Dynamic Range) scores, the Nikon D850 achieves a higher score at its lowest sensitivity (ISO) value. Once it exceeds this value, the Sony A7R II and A7R III sensors take the lead, delivering near-identical performance that is approximately 0.3 EV better than the D850’s. By ISO 25,600, however, this lead is extended to around a stop.

The Sony A7R III sensor maintains a high signal-to-noise ratio up to a sensitivity setting of ISO 800, whereas the Nikon D850 should be kept to ISO 500 or lower.

Conclusion: Outstanding image quality in the mirrorless segment

It’s clear that the Sony A7R III has a high-performing sensor that’s capable of capturing images with a broad range of color and tone, while keeping noise well under control.

However, comparing the A7R III sensor to the Nikon D850’s reveals the advantage that the Nikon camera’s lower minimum sensitivity (ISO) value brings. Photographers who predominantly shoot in bright light or capture motionless subjects with the camera on a tripod will record the most information, be it color, tone, or detail with the Nikon D850 set to ISO 32. However, if they require values above that, the Sony A7R III sensor produces marginally better images.

Sony’s in-body 5-axis image stabilization system is widely respected, and if it achieves the 5.5 EV shutter speed compensation in the A7R III that the company claims, it would enable the camera to capture images at ISO 100 instead of at ISO 3200 (provided the subject is stationary).

In this review, we have compared the Sony A7R III sensor to its most direct rival from a different manufacturer and to its predecessor from Sony. As usual, you can create your own comparison and in-depth analysis using our interactive image sensor comparison tool.

  • Hans J

    Does the a7riii still over heat under pro use?

    • J.R.Campbell

      ‘still’? There have been no reports of overheating that I’m aware of.

      • Andrew Garrard

        There were reports of the A7RII overheating I believe. I suspect that’s what Hans is referring to.

    • EyeBin Yerkinoff

      What’s “pro use” ?

      • Hans J

        Heavy wedding or fast-paced fashion shoots.

    • Molon Labe!

      There havent been any reports of the a7riii overheating at this point.

    • OSCE-NYSE General

      Like all Alpha cameras, there WILL be wither overheating, shutter failure (most common), firmware issues, and sensor banding. Literally not one Sony Alpha camera has been released without at least ONE of these problems, including the A9. Add to the fact there is one warranty repair center in the US and that adds up to a total joke of a camera. If you are looking to get one, I’d wait based off Sony’s plagued history of Alpha cameras.

      • Hans J

        Thanks for the well informed answer. I’ll skip this one again till Sony gets a good track record of making pro gear.

        • Don’t take that as gospel. In fact take that as trolling. There are overheating issues in 4K for many of their cameras but as Eamon Nerbonne below points out and provides a link to show that was resolved with the A7Rii and has not been reported with the A7Riii or A9. I get about 45 minutes out of my a6500 in Australian summer weather (and no limit on night shooting). Sensor banding is when using Electronic shutter under lights, this also applies to ALL cameras that use that. There are no credible reports of shutter problems and teardown show a very robust mechanism. Like I said: TROLL

          • Hans J

            Thanks for the additional information.. Sticking to Nikon for this generation till Sony sorts out all their problems. But what Sony has done for the market is great. They really have shook up the market. My new D850 is a better camera for it, thanks to Sony’s constant pressure.

          • By all means stick with Nikon, they are fine cameras but I question the phrase “all their problems”. All camera brands have had problems with individual models, Nikon no less than others—it has had recalls and ongoing issues, but that doesn’t make the D850 less of a triumph of DSLR cameras. It is a landmark; but so is the A7Riii and the trolling by OSCE-NYSE General is no more than that. There are plenty of quibbles about Sony’s choices and technical decisions but there are far more appreciative reviews especially of their latest offerings, the A9 and the A7Riii. In the end most of the quibbles come down to personal preferences rather than deficiences. See, for example: http://www.imaging-resource.com/news/2017/12/05/sony-a7r-iii-field-test-this-is-the-best-overall-professional-camera and https://www.dpreview.com/reviews/sony-a7r-mark-iii-review/11 where DPReviews final words on it were: “easily Sony’s best camera yet, and one of the best cameras we’ve ever tested. It’s the most well-rounded mirrorless camera on the market today, and for that, it earns our highest award.

          • Hans J

            All great things! and we all benefit from better and better cameras. I’ve shot Nikon for the last 12 years and none of them ever failed on me but 2 of my friends a7something failed. Just waiting for Sony to sort out the kinks and then maybe I’ll jump on board. Currently using a X-T2 for my small mirror-less solution, but those Sony’s sure do look nice.

          • Your story seems rather inconsistent. You start by implying you are interested in the Sony and yet, a little later, you tell us you already have the Nikon 850. You start by being “informed” of Sony faults and then, a little later, you tell us that you have 2 friends for whom “something failed” on their A7 Sonys—which models, how old, under what circumstances are all relevant here. You tell us that you are waiting for Sony to sort out the kinks—none of which are substantiated and yet you go stand by Nikon despite issues like this: https://petapixel.com/2017/07/14/nikon-recalls-d750-third-time-shutter-issue/. I have a lot of trouble believing you to be genuine. I have no problem with anyone choosing Nikon—or Canon, Fuji or whatever as we all need to choose against our requirements and they are not all the same, but your reasoning is puzzling. BTW for wedding photography Sonys silent shutter, eye focus and great video make a strong case.

          • Hans J

            If you read from the beginning I start by asking if this new camera has any issues just like all the other versions. All the Sony’s have has some problems. I’ve had two friends where their shutters stoped working during trips. Despite that when they Sonys are working they are amazing. I was hoping this new Sony would have sorted all that out… but now I realize my confidence in Sony is rather low and since I’ve shot with a Nikon D80 then D300s then D700 D800 D810 and now my new D850 that is only 2 days old, and non of them have had any issues for “ME” my confidence in Nikon is rather high. If this Sony proves to be a reliable camera, Next Gen I might go with Sony. Thats all I’m saying. You are the one that seams un-genuine do you work for Sony, all your post read like sales pitches.

    • gag21

      no sony camera will overheat in stills mode it only does 4k video for extended periods of time and that even has been solved quite a bit with firmware

  • Sasha Starets

    The Nikon D850 and the Song A7RIII are both fantastic cameras, no doubt. You could take amazing photographs with either. At this point choosing between the two will boil down to ergonomics, build quality, autofocus speed and accuracy, ecosystem, lens choice, etc. This level of competition is great for us.

    • If you need only a few static lenses, the Sony looks good.
      However, if you need a broad range of lenses, and shoot in the rugged outdoors, requiring long glass, the Sony-toy just has too many holes … and, really, the smaller-size as well as the shape is an impediment not a bonus.

      • Hmmm. I have over 150 lenses of many brands and every one of them work on my Sony A7RII. The number, types and brands of lenses able to be used on the A7 platform is one of its biggest strengths! With the battery grip installed the camera is not only easy to hold and flexible, it’s one of the most comfortable cameras I’ve every used for multi hour to all day carry.

  • Andrew Garrard

    I believe most of the ISO statements in the summary article are off by a stop. The minimum native ISO of a D850 is 64; the minimum native ISO of an A7RIII is 100, not 32 and 50 respectively as stated here (although it’s confusing on the graphs because DxO measures the actual ISO as lower than the stated one).

    If you actually go to ISO 32 “LO1” on a D850, unless it’s changed since the D810, you’ll lose a stop of shadow range – it’ll shoot at ISO 64 then divide the raw values by two (so you never get the highlight stops back). If the A7RIII has an equivalent ISO 50 mode, I suspect it’ll do the same. Obviously not so relevant in JPEG, but going below minimum ISO is generally a bad thing to do in raw – you may as well have the overexposed capture for the extra shadow detail, and pull it by a stop yourself.

  • Ruy Penalva

    Taking the fact the a7Riii has less pixels than Nikon 850, its sensor is clear better per pixel than Nikon.

    • Andrew Garrard

      True, although unless you’re actually trying to work out whether a resolution gain compared to another system is worthwhile to you, I rarely see the point in comparing the per-pixel values.

      • Ruy Penalva

        It’s a fault to not compare. Realizing that Nikon sensor could be Sony made, this is still more relevant. To me, given the fact a7riii sensor is stabilized, this is still more a gain not captured by the review. The identical score shows a common sensor maker.

        • Andrew Garrard

          Sure; it’s perfectly valid to point out that the Sony will have slightly better per-pixel behaviour – and the Nikon will have marginally better resolution. On the basis that the same total image is what’s usually relevant (and you’re not comparing, say, a 1000×1000 crop of each), I don’t usually find a huge amount is to be gained by looking at the per-pixel behaviour (some older full frame cameras look scarily competitive if you do…) – but it’s true that the difference is there.

          Each system has merits outside the sensor. Would I like Nikon to add the quad-exposure sensor shift and sensor-based IS? Yes. Would I like Sony to add an XQD slot in place of the UHS-I SD card? Yup. Battery life. EVF vs OVF. Ergonomics, menu structure, how responsive the camera is, video quality, AF behaviour… there’s way more to buying a camera than the DxO score. Otherwise, as I mentioned, Canon wouldn’t be doing so well. They’re both very good systems, and people should, if they’re lucky enough to be able to do so, pick whichever suits them. I’m still lusting after a D850 as an upgrade, but it’s much closer than Sony’s previous generation.

          • Ruy Penalva

            I’m talking about sensor and not whole system.

          • Andrew Garrard

            Sorry, I was picking up on your mention of the sensor being stabilised (which is a mechanical feature of the sensor mount rather than down to the sensor itself).

          • Ruy Penalva

            But this is a quality at sensor level, as such belongs to sensor.

          • Andrew Garrard

            I think I’m going to have to agree to differ with you about whether the mounting mechanism for the sensor counts as part of the sensor behaviour. It’s a valid and important feature of the camera (especially allowing for the multi-shot shift mode), but I draw the line at the actual conversion output from values received at the sensor. Otherwise we have to include things like shutter shock or EFCS. Once we get outside the basic conversion hardware, the line of what’s sensor and what’s camera is a bit arbitrary.

            And, to be honest, it doesn’t matter – people should consider the camera as a whole before they buy anyway.

  • slc320

    So dynamic range and color depth are better out of the Nikon, but only at Nikon’s the lowest native ISO (64)?

    • Andrew Garrard

      Yes (as has been the case for the D810 against various other bodies). The D850 holds up relatively well compared to its predecessors. On DR, it’s roughly a wash up to ISO 800, and doesn’t fall hugely far behind, though the A7RIII typically has the edge; the SNR score does favour Sony.

      I usually shoot at ISO 64 whenever I can, so I’m not in a hurry to sell my Nikon gear. 🙂 I can’t deny that the A7RIII has very nice characteristics, though. Both sensors took a step forward. Poor Canon (how will they cope with outselling everyone despite their lower scores?)

    • James Whitehouse

      No, only at ISO 32, if you look at the above graphs. It appears the Sony sensor has slightly higher DR at ISO 64.

      • Andrew Garrard

        Don’t misread the graphs – the article gets the numbers wrong. Nikon’s “ISO64” measures at (IIRC) 44, which is why the author got confused and claimed 32. Likewise the Sony “ISO100” setting measured low enough to be confused with ISO50.

        The nominal base ISO of the two sensors is 64 and 100 for the D850 and A7rIII respectively. Going under this with “LO1” on Nikon, or however Sony shows their equivalent, just throws away shadow detail for no extra highlight range.

        • James Whitehouse

          I didn’t misread the graphs. If the graphs are wrong, they’re wrong and DXO need to change them, but on the above DR graph the Nikon bests the Sony at ISO 32. At ISO 64, the Sony is slightly ahead (although by a hair).

          • Andrew Garrard

            There is no “ISO 32” on the above graphs, nor “ISO 50”. DxOMark plots the measured ISO position rather than the claimed ISO.

            If you look at the interactive graph you can see the lowest ISO figure on the Nikon graph is listed as “Measured ISO: 44 Manufacturer’s ISO: 64”, and the lowest on the Sony graph is listed as “Measured ISO: 70 Manufacturer’s ISO: 100”. (I do find it slightly odd that DxO always thinks the manufacturers under-represent the ISO, whereas at least on older cameras DPReview used to claim they tended to be spot on or generous. I don’t know whether sensors have changed or it’s just a measurement thing.)

            The article, which should be updated (*nudge*), talks about ISO 32 and ISO 50, neither of which are actually plotted on the graphs (it doesn’t say “44” and “70”, so it’s not trying to talk about the measured figure). I strongly suspect the author rounded the wrong way.

            You *can* make the D850 act as though it was at ISO32 with “LO1” extended mode; I trust you can make an A7RIII pretend it’s at ISO 50 equivalent in the same way. On the D850 (as with other Nikons) the extended mode simply records the frame at minimum ISO, then scales down the values in the raw file, which throws away a stop of shadow detail; you don’t get more highlight detail in return because the maximum representable value is similarly halved. It does affect the conversion to JPEG, but it’s not really useful for raw. The inverse is true for “HI” ISO settings – it uses the maximum native value, then scales the resulting recorded raw numbers up, which throws away highlights (because the values will max out) without gaining any shadow detail. So it’s unnecessarily confusing to recommend that people use “ISO32” on a D850 or “ISO50” on the Sony – they’re *not* “native”.

            If Nikon actually had an ISO32 native setting and it gained additional dynamic range, I’d use it, though. But we’re already getting close to the limit of Nikon’s 14-bit raw format.

            So yes, Sony has slightly higher dynamic range at its base ISO than the D850 does at equivalent ISO. In this case, both cameras are set to ISO 100, and DxO measures their effective ISO as 70. Not 64 (or 50), though, even though the points in the graph are closer to those values than 100.

          • James Whitehouse

            That’s a stupid way for DXO to present a DR graph. Very misleading unless someone explains it.

          • Andrew Garrard

            Assuming their measurements are correct (or at least consistently incorrect), it does allow comparisons where there’s a big difference between the effective ISOs (assuming you’re exposing equivalently). Why there’s such a disparity these days is another matter – dpreview used to claim manufacturers got it about right, but I don’t know whether sensors changed or if DxO just have an odd measuring scheme. (I’ve often wondered if they use a lens that blocks significant light somehow.)

            But no argument that it’s confusing!

  • Edward Lai

    A7rIII is a better all rounder at most on settings above ISO200.

    • OSCE-NYSE General

      Except when the shutter stops working.. or has firmware problems.. or sensor banding.. or overheating just like EVERY single Alpha camera released since the Mark I, including the A9. Photographers want a camera that actually… works, and one with a battery that lasts longer than 300 shots. A7s are a joke.. good for controlled environments like studios, otherwise useless in the real world.

      • Ioannis_Z

        Just clueless allegations without any fundamental prove.

        • OSCE-NYSE General

          Yeah, definitely lol specially when DPreview and FredMiranda were chalk full of complaints left and right of issues with their cameras. Just google it, why is it so damn hard to research. I’m not putting hundreds of links on here. Simple, google the fundamental PROOF, not prove.

          • Ioannis_Z

            DPreview and FredMiranda are not world instances. But I suppose you just google just around for everything. Why don’t you google the word RICK and add the first letter of PROOF so you can find the real meaning of your awkward name. Have fun with Google!

          • OSCE-NYSE General

            Well, if that made any sense I would understand it. No, I asked actual owners of the A7R Mark II if they would recommend it to me, and 40% of those I asked or checked reviews said no, where as 95% of D810/D850 owners said hell yeah.

      • Anulu

        You and your ridicolous statements are a big joke 😀

      • Bùi Hoàng Long

        No, you’re a joke. I had experience with Sony A6500, A7II and you can’t overheat these cameras just by taking photos! If you have problems with banding use the damn mechanical shutter (banding happens with all cameras that have electronic shutter)!

      • Edward Lai

        why does one need 12mp in a studio? Your camera shutter never stops working? I am considering the a7rII exactly because it is cheap. The shutter lasts for 500k cycles and i can use electronic shutter when I’m not using flash.

        • OSCE-NYSE General

          Yes, but WILL the shutter last 500k lol thats the issue. Owners were reporting on DPreview and FredMiranda that they were getting stuck with a bit of dust or cold weather. Errors. Freezing. Meh, no thanks

      • I am a Nikon 35 mm guy since the F2… but if I had the space and money I would have every single camera system that challenged the conventional wisdom of its time, from the first Leica’s to the current Sony’s (and a few Minolta underdogs in between). Don’t see the point on bashing makers or models… a lot of people used to do that with Minolta and it Rokkor lenses, would I have loved that Nikon had the foresight and cash to buy them then… Maybe and just maybe Sony would be buying the sensors from us and not the other way around, and probably we would be making better lenses than Sigma, Minolta was the underdog but their technology and lens design was second to none.

      • Alan Tunak

        I don’t know what experience or qualifications you have but as a professional i have a comprehensive kit D810, D5 and just about every top Nikon ever produced. I purchased the Sony 7r2 after trials and since it has become my choice weapon and now have some super lenses. Thousands of shots later it works like clockwork better images than the D810. Far less image noise, quiet in operation and not so heavy that you need to have a break every half hour like i do with Nikon. I do not know about sensor overheating as i do not work with video but have shot many images on a roll without experiencing that? The image stabilisation is wonderful and I make that my camera of choice. The batteries are not a major problem as i cary a bag of them just like sweets.They take seconds to change. I have just ordered the 7r3 just for the 2 card capability. Nikon images give a nicer skin tone but slightly veiled compared to Sony. Oh and i have had several Canons and Nikons repaired.

        • OSCE-NYSE General

          I’m sorry but I call b.s on “far less image noise” on D810 vs A7R-II. Literally, even at 12800 ISO, they are virtually the same, not even a stop better. Image quality, literally indistinguishable. The one thing you stated that is true is the weight difference. Otherwise, “image quality, noise performance” is a load of crap. The only noise improvement of an A7R II is in video, period. Next, the balance of the Sony is ruined because its small, so putting a 24-70 G lens throws the whole thing off. Professional? I guess that’s a matter of opinion. Fact is, the D810 remained the best landscape, high res camera until it got dethroned by the D850, otherwise.. the Sony’s 200 shot battery are laughable, horrible history of breaking down. The Alphas are great cameras, sure, awesome images and even better video cameras, but its not better than a DSLR.

          • Brian Kelly

            Man, I don’t know what era some of you guys jsut woke up in but your criticisms are either out of date or clearly untrue. Fanboys in the extreme.

  • VivaLasVegas

    It’s so hard to be a canon fanboy nowdays.

  • Mark Gasque

    Sitting over here with my inferior 5D mark iv while writing a deposit slip for my latest customers check. Very impressive technology though.

    • Andrew Garrard

      Quite. Any remotely recent camera is capable of taking good pictures – we’re down to minor differences (slightly better photos in ever more awkward conditions – although that may still apply to your once-in-a-lifetime shot). Sure, all else equal, go with the one that tests marginally better, but all else is rarely equal.

      I take a number of images with my D810 where I know the result would be, after local contrast adjustment, noisier with any current Canon sensor; I’m prepared to pay a premium for that behaviour, but it’s only one factor in the capability of the camera. It’s not like the Canons are worse than my D700, and I managed to take enough decent images with that (or it wasn’t the camera’s fault I didn’t). Canon sell more cameras than companies with “better” sensors; clearly other things matter more to many people.

      It’s only sensible to take the whole system into consideration when shopping (such as Nikon’s XQD support, Sony’s sensor shift, Canon’s dual SLR+PDOS autofocus). It’s good to have DxO’s measurements, but they’re only part of the equation.

  • Andrew Garrard

    It doesn’t appear that the ISO figures in this article have yet been updated, which makes me think that maybe Angela doesn’t get notifications about comments (or just has better things to do). I’m not on Twitter or Instagram, and wondering about this has led me to realise that there doesn’t seem to be an obvious email address for DxOMark (is there?)

    Any ideas how to get the editorial team’s attention on this, for the sake of posterity and potentially confused future photographers?

  • Yassine Eraman

    Waiting for the RX1R3

  • Thom von Schomburg

    Can you put a 13mm Nikkor Holy Grail on a Sony?? NO.

    Nikon F mount=Nikkor F mount lenses… ALL OF THEM.

    And, don’t give me your F mount adapter BS.

    Nuff said.

    • Andrew Garrard

      Well, you can adapt it; I’m not sure what’s “BS” about the adaptor in that case, since the 13mm is a manual-focus lens designed for film (not a sensor stack) anyway. While unusually wide in its time, it’s not so wide now (see the 12-24 f/4 Sony or the Sigma equivalent), and it’s not especially sharp (nor, sadly, is the 6mm f/2.8, which is also good for intimidating people). The Sony bodies seem to do pretty well with adapted EF lenses, including AF, albeit at a reduced shooting speed – you still get eye tracking for example. And, for “ALL OF THEM”, you can’t use an unmodified pre-AI lens on the D850 (and it has no full-time mirror lock-up for the invasive lenses even modified).

      So yes, when I get a D850 I’ll be able to stick my 200 f/2 on it and hit 9fps with the grip, which the Sony won’t do. And the D850 will write the results out to card somewhat faster (although it appears not entirely as fast as you’d hope given what XQD can do with a card reader). But Sony have a fairly decent basic lens selection (with a big 400mm coming), and better ability to mount third-party glass. It was easier to discount them a few years back – something I hope Nikon are aware of when and if they produce a (non-CX) mirrorless line.