Nikkor 500mm and 600mm f/4G ED VR lenses tested by DxoMark


DxOMark published their test results and conclussions for the Nikkor 500mm ($8,399) and 600mm ($9,799) f/4G ED VR lenses:

Comparing the 500mm with rival offerings, we can see that the Nikon is very close to the new 3460g $13,000 Sony. That said the Nikon is slightly sharper at full-aperture and has better uniformity, which is crucial in a lens like this and accounts for the slightly higher DxOMark Score. Compared with the newer, lighter (3190g) and more expensive ($10,399) Canon model though, it can’t match that at full-aperture nor has it the low levels of CA afforded by the three fluorite lens elements.

However, while the Canon 500mm mounted on 5D Mark III is sharper optically than the Nikon model mounted on D800, at the light levels used for DxOMark score (1/60, 150 Lux), the excellent dynamic range of the Nikon D800 sensor helps it improve the DxO Mark Score and accounts for the level-pegging. In lower light levels, the Canon would have the advantage.

Nikon AF-S Nikkor 600mm f:4G ED VR lens DxOMark test

Although the two lenses are extraordinary performers, the Nikon can’t quite match the new $12,799 triple fluorite Canon in sharpness or in lateral chromatic aberration, however overall the two perform very similarly. Both have homogenous sharpness at maximum aperture and possess low distortion and vignetting and excellent transmission, but reason why the DxOMark scores are the same is due to the excellent noise and dynamic range of the Nikon D800 sensor.

This entry was posted in Nikon Lenses and tagged . Bookmark the permalink. Trackbacks are closed, but you can post a comment.
  • HurrDurr

    Nikon dun goofd

  • Digital Keen

    All of you who think DXO is legit need to reconsider. They tie the Nikkor with the far superior Canon??? All because of the better DR of the sensor??? WTF? Seems like our government backs Nikon.

    • Dpablo unfiltered

      Don’t worry. They’re both fairly old lenses. In a couple years Nikon will have something that wipes the floor with your Canon lens, similar to their new 800, and they will still have better sensors. Your government is smarter than you and shoots Nikon somewhat exclusively.

    • Government? What does dxo have to do with government? dxo rates the system as a whole not the lens in isolation. They’re in fact explaining that the DR of the image the nikon system produces offsets the sharpness of the image the Canon system does (for the overall “score”). For that matter the d800 sensor is probably getting more resolution out of the lenses than the 5diii would so it’s nikon’s body vs. canon glass. If you only care about sharpness then the canon simply wins. Ignore the “score” — dxo breaks it all down explicitly.

  • EAJ

    So, despite its inferior performance in most parameters, the Nikon lenses receive the same score as the Canon counterparts because of the low noise and dynamic range of the D800. File this lens test under “bizarre”.

    • Nipon

      You need to think sometimes. 36MPx for the lens is not the same like 22Mpx.

    • zoetmb

      You can’t use the lens without a body, so it makes some sense. But since (I think) there are adapters to use Nikon lenses on Canon bodies, but not the other way around, they should have also tested the Nikon lens on Canon’s body. That would truly detail the differences between the lenses.

      • bossa

        To be fair and honest they should have put up the D3x scores rather than misrepresent the comparison.

  • Davidvictormeldrew Idontbeliev

    Seems bit strange to test the lenses on 2 different varying sensors – one on a mid high res 22mp body (5D) and the other on the highest FF DSLR 36mp D800.

    Would be better to test on flagship pro cameras the D4 and the Canon 1D X both of similar mp’s.

    • Dpablo unfiltered

      The Nikon would still have superior dynamic range.

  • reader

    What braindead comments! DXO clearly states how the D800 _combo_ gets the same score despite lower maximum rating. Learn to read before posting!
    And yes, it is the _combo_ that give you pictures.

  • Rick

    well that really sucks that nikkor super tele lost out to canon’s….well..if there is anything, the nikon is a few Gs cheaper in all the super teles lineup

    • Dpablo unfiltered

      Yes, and the new 800 is sharper than ALL of those lenses, probably even with the converter on it. And it’s not much more money than the competing 500 and 600mm lenses. So I wouldn’t be worried much, you’re comparing some fairly old glass with some that’s about new and priced higher too. They’d damn better be a little sharper…

    • Jon Rista

      Canon’s list prices are not necessarily what one pays. I just picked up a brand spankin new, with all accessories and the hard case, EF 600mm f/4 L IS II for $10,856 USD. That is only $1000 more than the Nikkor EF 600mm f/4 ED VR II, which has had a relatively stable price around $9800 USD (for non gray market versions) for some time.

      In terms of lens design, there is no question it is a superior lens. If you could find a way to slap it on a D800, it would trounce anything Nikon has to offer with the possible exception of the $18,000 (!!!) new 800mm, which actually borrows design principals from Canon with its new fluorite elements (and yet, still ends up heavier.)

      • Rick

        Well but actually I can get the nikkor lower than what you are saying. I got a 600mm vr like new with all accessories for $7300 cash, so the staple market price of 9800 doesnt really apply in some case. So there, 7300 is still about 3k cheaper than your Canon
        For super teles, one tend to get good deal because not many shooters need lenses like this even if they can afford it

        • Jon Rista

          “So there” LOL, sorry, got a kick out of that.

          So, assuming anyone can pick that lens up for that price these days (I’ve havent seen it less than $9300 for over six months, and that was gray market), the real question is does one get real value for that extra $3000?

          I could have picked up an old EF 600/4 L Mark I…and dirt cheap, too…for ones with minor cosmetic scratches on the tripod foot, one can get it for about $6500. I explicitly chose the Mark II, for the superior IQ for sure…but far more importantly for the incredible weight savings. Your $7300 600/4 ED VR still weighs an astonishing 11.2lb! That takes it well out of the range of hand-holdability, especially when you tack on a camera body, probably a TC, and likely a battery pack.

          I may have spent a few grand more on my lens, but I am by no means complaining about it! I can hand-hold my 8.5lb 600/4 for hours, making it an entirely viable tool for both BIF as well as wildlife on the move.

          The question isn’t who’s cheaper…it’s whether you get realizable value for what you pay for. I not only saved over two grand off the list price…I got a brand new lens that offers superior IQ (with a TC attached) that weighs three and a half pounds less than the far cheaper alternative, giving me far greater flexibility in the field (flexibility which doesn’t require me to be a creatine-munching, testosterone-driven muscle monstrosity).

      • Birdshooter

        (and yet still ends up heavier?? where do you get this???The Nikkor AF-S 800mm f5.6E FL VR lens weighs 4,6kg – almost exactly what the Canon EF 800mm f5.6L IS lens weighs. I had the Canon and I now have the Nikon – this lens is in a whole new class of its own re sharpness and also VR (rated at 4.5). I shot a handheld image 1/50th second that came out sharp….

  • bossa

    The Nikon 500/4 is ‘only’ 15 perceptual MP on the D3x – 5MP LESS than the Canon on a similar resolution sensor. Not a very flattering comparison although the canon does have those extremely expensive Fluorite elements.

    • bossa

      ..and the ‘rating’ for use on the Dx3 is 21. That puts it in another ballpark, but so does the price of the canon.

      • bossa

        Sheesh! It’s great start to the day for me… I meant to type D3x

    • bossa

      My bad: The Canon is 19PMP.. not 20 as I stated above. Still, the Nikon lags by 4PMP on a similar resolution sensor. I wonder how much damage the flat protection ‘element’ does to the figures.

    • fjfjjj

      …and when you find out what a perceptual MP is, call me.

      • “The number of P-Mpix of a camera/lens combination is equal to the pixel count of a sensor that would give the same sharpness if tested with a perfect theoretical optics.”

        • fjfjjj

          I’ve read this link. There is no solid description of how the measurement is made. It is not reproducible by a third party. Not buying it.

          • You’re being disingenuous then since you say elsewhere the measurement was “pulled out of their asses and never explained”. You just don’t like the explanation.

            “DxOMark’s new Perceptual MPix measurements are based on acutance and human contrast sensitivity function (CSF) published in recently-released image quality standards from the International Standards Organization (ISO) and the International Imaging Industry Association (I3A).”

            Here’s a paper on how they developed their spatial metrics:


            I haven’t tried to read and comprehend all the nitty gritty, but i’d at least try to before making dismissive comments — repeatedly — with no foundation.

            • fjfjjj

              Thanks for the link to the paper, however I will note that it does not address “perception” in any significant way.

            • So, basically, you just keep moving the goal posts so they can never meet your requirements. I understand.

            • peterw

              well, the meaning of the word ‘perception’…

              … one could consider that ‘perceptual’ too…


              I guess, in the end Tonio and you both think the same about DxO tests.

              I have the same experience as Tonio with, very useful and coherent with my own findings.

              Valuable discussion… thanks both of you.

            • Well, that’s the thing, DX0 makes statements with no reliable foundation other than black box numbers. Their referencing ISO standards does note completely explain where their numbers come from. One doesn’t have to comprehend fabricated statements to recognize they are fabricated. fjfjj is on solid logical ground here.

              The best tests you get are from DPR and IR, who make an attempt to test under controlled conditions. They explain their test setup in detail. Sure, they make errors sometimes, but they’re the only ones with any useful data. Which happen to be files directly from the cameras. That’s the only thing that really matters in the end anyway.

              All DX0 can tell you is how their software can handle files from the cameras. Which is only important if you’re interested in their software–and boy do they want you to be!

              In the end, what matters more than any of the measurebating is: how do the files look. And that can only reliably be tested by YOU with your lenses and your camera in your hands.

              Anything else is arguing to argue.

            • I like DPR as far as it goes, but shooting a scene in studio lighting only gets you so far (high ISO in crappy lighting is very different from high ISO in good lighting, focus on a test chart is different from focus at infinity). I also like that DPR seems to be taking more sample shots in poor light. I think Photozone does a better job with lenses in many respects by being reasonably consistent in its real world test shots which include things like shots focused at infinity and so on.

              All that said, dxo’s measures are pretty transparent, and its “perceptual MP” is a pretty straightforward concept — how much detail are you getting with a given lens/body combination. The math behind what they’re doing is complicated but it’s not BS, and it pretty much matches my (limited) experience. Insofar as I have lenses they’ve tested, the lenses they show to be soft are and the lenses they show to be sharp are.

              I think you need to take dxo for what it is — a company that sells photo software and is pretty good at measuring distortion, vignetting, etc. and adjusting for it. Their ISO number is voodoo except insofar as bigger is better. Their T-stop number on the other hand is probably simply correct, as are their distortion and vignetting numbers.

              The problem is the overall score which is what people argue over.

  • Just saying

    Nikon guys are paying less for both their camera and lens and getting the same image. No wonder Canon guys are upset. They got taken. I think Sony guys all fainted.

  • Smudger

    The Nikons would have been even sharper if they had been fitted with an RRS or Wimberley foot instead of the standard abomination.

  • itsmyname

    Do you really care what DxO spills out ?
    THEY are flawed in many areas its not funny.

  • waterengineer

    Were the Nikon images reduced to the same resolution as the 5D for comparison? If not, I bet the Nikon actually scores higher.

  • fjfjjj

    What’s a perceptual megapixel (P-Mpix) again? Oh yeah, something DxO pulled out of their asses and have never explained.

  • VivaLasVegas

    36mp vs 22mp…….WFT, Nikon is getting BIG TIME a$$ whoppin from a smaller MP body, stick a D600(24mp) into that Nikkor and the sharpness score will be down a 13P-Mix, ah ha ha ha consumer grade Canon lens!

    FACT: 6D(20MP) is a better match up versus D800/E 36MP body in resolution test!

    • Dpablo unfiltered

      Really, I would expect either lens to give about the same resolution regardless of the camera unless you were to go back to using something like 12 megapixels in which case it would be limited by the sensor. And D800 vs 5D is appropriate because they are the cameras that both have in that price. But go ahead and shoot a nasty plastic 6D if you like. I don’t know if you could tell the difference.

      • Neopulse

        Actually I owned temporarily a 5D MKIII and I’m currently selling it on an ebay-like website where I live. It wasn’t as solid as a D700 or D800 in the hands. Felt cheaper, although selling it for different reasons and not for the build. Gonna start saving up for better gear.

    • Maji

      TROLL ALERT!!!!!!!!!

  • Brian

    Is this a lens review or body review?

  • Eric Calabos

    500mm f4G on D3s > 8 P-Mpix
    50mm f/1.8D on D3s > 10 P-Mpix

    am I only one who feels its ridiculous?

    • Dpablo unfiltered

      Wide open. Wide open…

  • neversink

    So maybe my 500mm is better than the Canon 500 mm if I shoot it on my D800, What if I shoot it on my D4? Do I really care what DX0 says. All I know is that the photos I take with this lens and any Nikon fx body are incredibly sharp. As I am sure the Canon 500 is sharp on any fx Canon body. I am happy with my images. So, do I really care what Dx0 says in the long run? No, but it certainly is interesting. I’ll keep using the Nikon 500mm no matter what Dx0 tests state.

    • Sleeper

      You cared enough to come here and QQ about it.

    • jk

      well, I think in this area Canon is a bit better but just because it is the latest of its kind out.

      so next 500/4 Nikon comes , it will change again.
      anyway , they are keep leap frogging every gen to come.

  • Martijn

    “at the light levels used for DxOMark score (1/60, 150 Lux)”

    Who shoots a 500 or 600mm at those speeds?

    • Someone doing landscape. Which are also the people who can close down a little and make the differences disappear.

      Otherwise you’re right–usually these are used for moving targets

      But I wouldn’t waste too much clock time worrying about the latest garbage DX0 has made up to get attention.

  • NikonFanboy

    I think we are comparing 2007 Nikkors against the latest 2012-13 canons. Its not fair. We need to wait for the Nikkors 200/300/400/500 adn 600mms redesigned with FL elements. With new FL Ed elements there will be considerable decrease in weight and size. There should also be gain in performance and the resolution should match the canons. We will just have to wait for Nikon to release the new design. If they get 400mm any where the weight of canons i will be up for it;0

    • jk

      right , but why do they not compare the new VR2 version to the IS2 Canon?
      and I think in any other test I have seen the Sony comes up better than the other 2.
      so it is hard to trust DXO when it comes to high end lens test like this.

  • Graham

    Why not just fit the Nikon lens on the canon body and measure…….. The measurement method here is ridiculous.

  • Spy Black

    Boy do you love to rile up the natives Admin…

  • Gary

    Now People…give me any one of those lenses and any body and I’ll outshoot any of you. The differences are that small, and yes, I am that good. [VBG]

  • MB

    So now we know, Nikon lens sucks …

  • FarQinell

    Unfortunately they never tested the tripod mounts… Nikon’s are nearly always persistently crap.

  • Jacob Wadsworth

    I have always been a Nikon fan. I have never changed ever since and I never regretted my decision. Nikon is user-friendly at the same time that it is cheaper than most popular camera brands. The quality of the photos are superb too that is why people prefer using the Nikon brand. –

  • peterw

    very strange that the camera is involved in rating a lens.

    I doupt if it will be visible in pictures, until the moment a convertor is applied. Probably the Canon lens will accept convertors much better.

    Compliments to Canon…

    I rember how perfect the 600 mm F5,6 ai-s IF-ED seemed to be, until I got my first version of 500 F4 AF-S… ‘autre chose’…
    … which didn’t have VR while many a Canon bird shooter had a 500 F4 IS about eight years before…

    VR improved increased the number of sharp shot (of sitting birds) slower than 1/200s by some important factor. The new VR version seems sharper than the first AF-S I had, but there was a sort of tendernes in the sharpnes to the first version (I don’t mean bokeh which has improved in the VR version). It’s the same with the new 50 F1,4 and 85 F1,8 AF-S versus the AF-D versions. The newest are tack sharp, but that’s not always what you want…

    The good images of the 600 F5,6 are still usable :).

    (how come the much praised Sony comes out so bad? anyone?)

  • InFocus

    were at Nikon here – but sorry – the weight is what counts if you travel –
    so i choose Canon for Supertele.
    And I’ve testet 2,8/400 – Nikon against Canon – Canon is sharper at f:2,8 and much better with extender, and much lighter – thats what counts in the field.

    Camerabody doesn’t really matter – if I need a 5,6/80-400 mm – I would buy a D7100 – I always have Nikon and Canon bodys.

  • Mikael Risedal

    The 500 and 600 are inseparable from Canon, Hasselblad MTF lab in Gothenburg have meassured all super teles from Canon, Nikon, all lenses has a optimum regarding distance and DXO are meassuring one distance.
    Can I post the real MTF somewhere?

  • Mikael Risedal

    Fluroit or no fluorite, the Nikon 400/2,8 has less CA than the new 400/2.8 from Canon

  • Back to top