Sigma 24-35mm f/2 DG HSM Art and Tamron 18-200mm f/3.5-6.3 Di II VC lenses tested at DxOMark

Sigma 24-35mm F2 DG HSM A lens review DxOMark
DxOMark published their test results of the Sigma 24-35mm f/2 DG HSM Art lens for Nikon F mount ($999):

All three models perform well, with Nikon’s own Nikkor brand 24-70mm f2.8 coming in third with a DxO Mark score of 28 points, and the Tamron in second place with 31-points, but that’s not such a surprise, given the extra level of difficulty in making wider-range zooms. Still, it does show just how good both the Tamron and Sigma models are, and the latter’s wider aperture is at least as challenging optically as the extra reach.

Sigma’s higher peak sharpness and low levels of chromatic aberration are impressive, but at f/2 the Sigma isn’t as sharp as either the Nikon or Tamron at f/2.8 over the same focal range. With the Sigma stopped down to f/2.8, though, it is superior to both, and particularly so in the outer field at 24mm. Stopping down further, the Nikon doesn’t fare that well in the outer field due to a combination of strong chromatic aberration, curvature of field, and astigmatism. The Tamron comes close to the Sigma from f/4 through f/8, but similarly suffers from some astigmatism, which can be troublesome (although it depends on the subject and its position in the frame).

All three models have better control of uniformity and fringing at 35mm, but the Sigma leads until f/8, where the Tamron matches it more or less across the whole image field. As for vignetting, at maximum aperture the Sigma has more to contend with than its rivals, and it has slightly higher pincushion distortion at 28-35mm, but it has lower barrel distortion than either at 24mm.

At around $999, the Sigma 24-35mm f2 DG HSM A is reasonably affordable, fast, and optically outstanding. It can easily take the place of a 24mm and 35mm lens, but of course it is not as convenient as the more typical 24-70mm focal range of standard zooms. And as we’ve seen recently, manufacturers are adding stabilization to those models, making them a lot more attractive to a wider audience (if you include videographers). Nevertheless, the Sigma appeals to same crowd and is more accessibly-priced than either the new Nikon or Tamron. In spite of the somewhat narrow focal range, it is an attractive alternative, and it will likely be a popular choice with enthusiasts looking to cut down on the number of lenses they carry.

DxOMark also tested the Tamron 18-200mm f/3.5-6.3 Di II VC lens for Nikon F-mount ($249):

Tamron 18-200mm f:3.5-6.3 Di II VC vs. Sigma 18-200mm f:3.5-6.3 DC Macro OS HSM vs. Nikon DX VR 18-200mm f:3.5-5.6G IF-ED lens comparison

Tamron 18-200mm f/3.5-6.3 Di II VC vs. Sigma 18-200mm f/3.5-6.3 DC Macro OS HSM vs. Nikon DX VR 18-200mm f/3.5-5.6G IF-ED

Tamron 18-200mm f:3.5-6.3 Di II VC vs. Nikon AF-S DX 18-140mm f:3.5-5.6G ED VR lens comparison

Tamron 18-200mm f/3.5-6.3 Di II VC vs. Nikon AF-S DX 18-140mm f/3.5-5.6G ED VR:

DxOMark's conclusion - good value for money:

Lightweight, convenient, and affordable, the Tamron 18-200mm f/3.5-6.3 will be a popular mass-market lens for enthusiasts shooting APS-C. Covering a versatile wide-angle to long telephoto focal range in a relatively compact and lightweight unit makes it challenging for super-zooms to deliver top image quality however. As we’ve seen, the Tamron 18-200mm f/3.5-6.3’s overall DxOMark score of 13 point ranks towards the lower end of the performance spectrum when tested on the Nikon D5300. Within its own APS-C super-zoom 18-200mm class, it achieves the same overall DxOMark score of 13 points as its Sigma and Nikon rivals. Both the Sigma and Nikon versions offer slightly improved sharpness, however, and offer better resolution than the Tamron at a 50mm focal length. At 200mm, sharpness for all three lenses isn’t great, however, so if it’s good performance at long focal lengths you desire, you should consider a lens with a shorter focal range such as the Nikon 18-140mm. With the addition of Tamron’s Vibration Compensation system on the new 18-200mm f/3.5-6.3 super zoom, however, together with the weight reduction down to just 400g — around 1/3 lighter than the Nikon alternative — it’s a good value-for-money option for those photographers looking to lessen the load.

This entry was posted in Nikon Lenses and tagged , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.
  • D600spots

    DXO mark says the Tamron has the same vignetting as the Nikon and the Sigma has more? Most reviews I have read have complained that the Tamron has more than the Nikon. What’s the real story?

    • Eno

      I only shot with the old Nikon 24-70mm f2,8 and compared it to the Tamron, but as you said, the latter has more vignetting.

  • Photobug

    Good update. It’s healthy to see Sigma and Tamron releasing good glass. That will continue to push Nikon to make better glass. Still prefer Nikon over those other two from a “build quality”. Too many friends have had their Tamron and Sigma lens fail just outside of the warranty. Yes, there are exceptions…just saying that is my friends experience with Nikon and Canon lens mounts.

  • I’m not sure about the credibility of anyone using the word “convenient” in the same sentence with a lens that’s six inches long and weighs two and a half pounds.

    • Patrick O’Connor

      My station wagon is a lot more convenient than the motorcycle I had several years ago! :-/

      • Spy Black

        Yeah, but which was more fun?…

        • Patrick O’Connor

          Not even close! 🙂
          Although, there was this time with a young lady in a station wagon I had even before the motorcycle…

          • Spy Black

            Well, at least I still have my motorcycle. Could use another young lady… 🙂

            • Patrick O’Connor

              My wife is 2.5 months younger than me so I’ll ALWAYS have a young(er) lady! 🙂
              What kind of bike do you have?

            • Spy Black
            • Patrick O’Connor

              Not what I had imagined at all. This looks like what a 20-something would ride and I had the impression you were a bit older.
              Even when I was young, I couldn’t fold myself onto one of those things. 🙂
              Very nice!

            • Spy Black

              That’s actually a full upright seated bike, better known as a UJM (universal Japanese motorcycle). If you rode a Harley or other cruiser type, everything else looks like a sport bike. :-). Quite comfortable all day rider, even for a 60-year-old like me. It only does about 135 mph. 😉

            • Patrick O’Connor

              Pff! My station wagon goes faster! 😉
              Seriously…it does.

            • Spy Black

              I told you it only goes 135. 🙂 It’s only a 650. However 135 feels a bit differently on a bike than in a station wagon. 😉

            • Patrick O’Connor

              Actually, I’ve never driven that fast in, or on, anything. It’s more of a, “I could if I wanted to” thing. I really don’t want to!

            • Spy Black

              So how do you know your station wagon goes faster? 😉 I’m really a backroads rider. Anyone can twist a throttle on a highway. It’s the twisties that separates the men from the boys.

            • Patrick O’Connor

              The speedometer goes to 160. I’m guessing it’s not a boast but maybe it is!?
              I loved riding on backroads but I got a nasty scare once so I learned a healthy respect for them. This was many, many years ago when I was young and stupid. I’m not young anymore! 😉

            • Spy Black

              That’s quite the boast. LOL!

            • Patrick O’Connor

              “Beast” is very accurate. You can almost see the gas gauge dropping even if you’re driving conservatively.

    • JJ168

      Hmm, i think it is convenient when considering (IQ a side) to bring a sigma 24 35 and 50 Arts which combined weight more than the nikkor and changing lenses, missing shots and so on.

      • Patrick O’Connor

        And, you lose the 70mm end.
        Of course, Pete could be comparing the new Nikkor 24-70 to the Tamron which is more svelte.

        • My Hodaka only weighed 195 lbs!

          • Patrick O’Connor

            That’s kinda heavy for a moped, isn’t it? 😉
            When my Japanese nephew visited us a few years ago, he was pretty proud of his 400cc motorcycle. When I told him it was kawaii (cute), he wasn’t very happy. 🙂

            • You’re just a shit stirrer, aren’t you? 🙂

            • Patrick O’Connor

              Actually…yes. I think people are FAR too serious about stupid stuff (like photography gear for example) so I try to use humor to break things up a bit.

      • For fashion work I’d agree that this lens is convenient and provides the necessary IQ that that work demands. If you’re doing an energetic shoot with lots of movement and you want that spontaneity, changing lenses is a cold shower for sure. I wonder about the journalist types, though. I see them with these lenses and can’t figure how it makes that much difference by the time it gets published. I mean, the 24-85mm f/2.8-4 weighs a fraction and the IQ is excellent. One stop difference doesn’t give you THAT much less subject separation. Durability? If you break the 24-85mm, go get another one!

    • Spy Black

      I think the problem is just comparing apples to oranges. The 24-35 f/2 is a significantly different beast that the 24-70 f/2.8 lenses. Although the Sigma is an interesting lens it would be great if they updated their 24-70, which has always been a crappy lens. The release of the 24-35 may have short-circuited that.

      • I was joking…there’s no comparison, and I didn’t make one. The 24-70mm is a big, heavy lens and I don’t see anything “convenient” in that…other than having all those focal lengths in one place. I’d bet it weighs more than the 24 f/2.8, 28 F2.8, 35 f/2, 50 f/1.8 and 85 f/1.8 combined. Or close anyway.

        • PhilK

          The new 24-70mm is outrageously large and heavy. It’s longer and far heavier than the 70-300VR, heavier than the 70-200 F4 VR, and within a couple hundred grams of the 80-200 2.8D. I find that kind of outrageous, personally.

          And with a long side of 70mm, to me this is more of a ridiculously large and heavy wide-angle zoom than a walkaround lens. I have a hard time understanding the appeal of it.

          • Mike

            You should try it Phil. What they don’t say about the weight is that it feels rearward biased. That if you rested the middle of the lens on a finger, it would actually tip back. When mounted on a camera that rearward weight distribution feels great. The camera/lens combo feels balanced. conversly, as sharp as the new Sigma’s are, they are quite heavy for their focal lengths and feel far more front tippy. Again, something numbers can’t measure.

            Perception is funny. If you can handle a 70-200 2.8 or heck, the f/4 version, all day, the 24-70 is no problem either.

            • KnightPhoto

              Agreed in general on the 24-70E form factor. I brought my G down and compared, and on a BlackRapid strap I’m not going to notice weight diff and size diff is negligible.

              Do you notice the “less sharp in centre” or is that simply a bench-test phenom?

            • Mike

              https://500px.com/photo/137088477

              2.8 at centre is pretty good for me. 🙂

            • KnightPhoto

              Very nice Mike! Thank you,

            • PhilK

              I just am constantly amazed at how gigantic modern lenses are compared to the old MF or even early-AF days. Yeah, they have AF and some of them have VR. But Nikon’s lenses in particular seem to be some of the biggest on the market for FF lenses. FFS, the 24-70E is within 10% of the weight of my old 400mm ED-IF (AI-S) prime, including its tripod foot!

              I discovered one explanation when I read lensrentals teardown of the non-VR 24-70, which among other things compared it to the Canon 24-70. The LSI chips in the Nikon were so large, that unlike Canon and increasingly most of the 3rd-party lenses, instead of a nice clean PCB on the back around the rear lens group that had most of the electronics on it, Nikon had to put boards all over the sides of the barrel, with flex-circuits and wires running all over the place. (Which incidentally also makes it harder and more expensive to repair)

              I’d love to see a teardown on the 24-70E to see if they improved any of that – tho judging by the immensity of the VR version, I’d guess not.

          • The irony in my comment is that, if you’re a photojournalist, the lens IS actually convenient in terms of getting one lens with all those focal lengths. Journalists and some of the fashion shooters like the range and speed and just put up with the size and weight. But for everyone else the thing is just ridiculous. I’ve done some political campaign shoots that wanted that “newsy” look for their promotion and have rented the old style 24-70. By the end of the day I’m whipped from hauling it around…but I’m an old fart so, you know, everything hurts.

            • PhilK

              Well I’m no spring chicken any more myself, but at least if the thing had some long-end reach I might be inclined to consider it… but to me, 70mm is too anemic to justify a lens that is bigger and heavier than a ~200mm prime.

              Nikkor 180mm F2.8 AF-D:
              weight 760g
              length 144mm
              width 78.5 mm

              Nikkor 24-70 F2.8E AF-S VR
              weight 1070g
              length: 154.5mm
              width 88mm

  • Mike

    Had the Tamron, sold it and bought the Nikon 24-70 VR. All I’m gonna say is don’t knock til you try it… for more than 30 minutes at a store. Built better, supported better, far far better AF, and optics are excellent. DXO can’t measure the intangibles. And a number score doesn’t tell the whole story.

    • Eric Calabros

      Its clear to me DXO tested a bad copy, but they never admit it.
      Field maps are from DXO, corner samples are from Cameralabs, both 24mm@2.8. It’s like I placed it wrong, but I didn’t. It shows exactly why you should never trust a single review test:

      • Spy Black

        lensrentals.com is an excellent source for checking lens performance, as not only do they average their results from 10 or more copies of any given lens, but they test them on an optical bench, which removes a camera body’s influence on results.

        • PhilK

          Their test of the 24-70E was interesting, basically concluding it gives up a little in the center but is the best on the market on the edges.

          I sure hope so, for that monstrosity..

        • Eric Calabros

          people say they test at infinity on bench, so real world result might be different.

          • 24×36

            I would guess my shooting is closer to infinity or hyperfocal than it is to close focusing distances most of the time, so bench testing may actually be a better indicator of performance from the perspective of how I shoot. YMMV.

        • 24×36

          The multiple samples type of test is definitely advantageous. However, I don’t necessarily think that camera body testing is bad, as long as you can compare on the SAME camera body (so it isn’t an apples to oranges type comparison).

          Nice thing about DxO is they keep updating their tests of older lenses by testing on the latest bodies, so you can do apples to apples comparisons with newer lenses. Most testing sites only test on whatever body they were using when a given lens was first introduced, which makes it impossible to do a meaningful comparison of older lenses with newer ones.

      • PhilK

        All I can say is with chromatic aberration fringing measured at 30um, it looks like something is wrong with the lens.

        Is Nikon really relying that heavily on in-camera chromatic aberration compensation that they really design lenses like that now? Trying to use a lens like that on film would be a fringy disaster. And isn’t the in-camera correction only applied to JPEGs anyway? (Otherwise DxO wouldn’t have been able to measure it, I assume.)

        • Eric Calabros

          lets see what Sony will be able to do with CA in its upcoming 24-70 f/2.8.

          • Patrick O’Connor

            Just curious why that would mean anything!? If you’re looking at lenses for a Nikon, it wouldn’t matter if the Sony is perfect or pure crap. If the Sony were a lot better, can you take it to Nikon and demand they match it? I’m not trying to be a jerk (it comes naturally) but I’ve never understood comparisons for incompatible gear.

            • Eric Calabros

              It doesn’t change the situation for other brand users, it just can be labeled as a proof that “making X improvement in Y optical solution, is not affordably or conveniently achieveable with current tech, otherwise engineers from other companies would make it possible”.

            • Patrick O’Connor

              Again…to what benefit.
              There is a plethora of proof that anyone can become a great photographer but it does me no good whatsoever! 🙂

            • Eric Calabros

              There is a benefit in making a conclusion that waiting for next iteration of the lens is wasting time because its unlikely that desired improvement appears soon.

            • Patrick O’Connor

              I never think about stuff like that unless a new version is imminent. I’ve always heard that ‘When you need something – buy it. If you don’t need it – don’t’ and have never seen a reason to do otherwise.

            • PhilK

              I think Eric has a valid point that relates what is realistic to expect Nikon to produce. We’re all fanboys(girls) to some extent, but if it gets to where Nikon consistently falls short of what is available from the competition, it starts to undermine their customer’s loyalty and pushes them towards switching.

              Case in point: D500. Nikon surely lost a significant number of serious amateur and pro users because they had a gap in that category, despite users begging them for something to replace the D300s for 4-5 years now. If we all just accepted Nikon’s rationale that the D7xxx series was supposed to be the replacement (and if the 7D-II didn’t happen), we’d still be waiting.

              It took close to 4 years from the D300s introduction before the “replacement” (D7100 in 2013) had an AF system as good as the predecessor D300 from 2007.

            • Patrick O’Connor

              And it’s a good thing they finally came out with the D500. You couldn’t take a decent photo with the D300s if your life depended on it! :-/

              The only reason I’m on this site is to see what is coming out and plan future purchases. But if Nikon et. al. never make another new product, I’ll happily shoot with what’s already out there.

              What is realistic to expect from Nikon is they’ll do whatever is in their best interest as a company. In some cases, that’ll make photographers happy and sometimes not. And that’s all I have to say about that.

      • TheInfinityPoint

        I totally agree. I compared the E version (rented) against the G version on my own, and the corners are worlds better on the E version.

        I trust DxO less and less these days. Unfortunately, the naive fanboy will inevitably quote DxO scores as truth.

      • 24×36

        Same can be said of all lens tests. Sample variation is always an issue, tests of a single lens are always just that – a test of a single lens, and while they may be representative of a certain lens as a group, there is always a chance of an outlier sample (one way OR the other) and such “tests” should therefore always be taken with a grain of salt to some degree.

      • PhilK

        This dovetails with a review I read recently, and you can see it right in the MTF curves for the E. It has this weird U-shape in a bunch of the curves where sharpness declines slowly from the center to middle (nothing unusual there), but then does this weird about-face and skyrockets up as it goes towards the corners again. Nikon apparently really wanted the corners to be sharp on that lens.

    • Eledeuh

      I trust Nasim Mansurov and his preview on this one, in short he observed that the lens is sharper on average, much sharper in the mid-frame and corner parts, and a tad less sharp in the center.

      The new 24-70 f/2.8 VR has gained a lot of appeal for landscape photography, and maybe lost a bit for portraits.

      • Mike

        I do too. I even made a comment on his review to some of the detractors. I shoot weddings, and I don’t think I have ever or would ever shoot a wedding with people dead centre for the entire day; getting ready to reception dancing. I shoot a lot off-centre. Like, a lot off-centre. Where you might see a lens’s weakness in terms of vignetting or sharpness. The new Nikon VR has made those concerns go away. I use primes yes. But this zoom is so good. Far more reliable on the edges than the Tamron 24-70. Sharper and brighter.

        I don’t know why dead centre is even a metric when it comes to lenses because even with portraits in portrait orientation you’re using the upper 1/3, at least, to focus on eyes.

        • AYWY

          Great points. When people harp on about center sharpness, it’s like they never went through photography 101 lessons on rule-of-thirds and golden ratio…

      • Patrick O’Connor

        Nasim’s testing methodology appears to be sound but his conclusions are suspect. I’ve seen him pan lenses that did well in HIS testing and rave about lenses that were only meh.
        That matters to me because I’m not nearly as technical about this stuff as a lot of you guys and usually jump to the conclusions first. If I have a spare moment, I put it toward improving my photographic abilities (or maybe it’s “lessening my inadequacies”) rather than trying to understand the tech stuff. If I, or my client, can’t see it in the photos, it doesn’t matter.

        • Eledeuh

          > rave about lenses that were only meh

          Which one are you thinking about ?
          I’ve seen many more people fail at reading what he writes than him being outrageously biased in one of his reviews.

          • Patrick O’Connor

            Sorry. It’s been a while and I can’t remember.
            As for failing to read what he writes, I can’t say that I’ve always read every word. Again…I tend to jump to the conclusion and then backtrack if I think it’s warranted.

    • DXO’s lens tests just make no sense. For example, their score says the Sigma 24-70 2.8 EX is equally as sharp as Nikon’s 24-70 2.8 (non VR). Anyone who has used the Sigma knows it’s an optical disaster, nowhere near the great quality of the Nikon. What even disturbs me more is when sites use these weird numbers to create unnecessary drama, like this: https://fstoppers.com/originals/pixel-peeping-reviews-nikons-24-70mm-f28e-ed-vr-leave-room-sigma-kill-102375

      DXO’s lens tests are probably the worst thing that has happened to the photography equipment community…

      • fanboy fagz

        Yes the sigma 24-70 is garbage. No contrast or sharpness till f/5

        But that 24-35 art rapes them all. I cant wait to see what sigma art will show up this year.

        I always say that dxo is full of shit

        • Spy Black

          Don’t know if the 24-35 precludes it, but I would like to see an updated 24-70 f/2.8 in the ART series.

          • fanboy fagz

            if you look back on every single 24/28-70 2.8 they made, they all were crap. one after the next. Id also like to see a 24-70 art. not sure they can do it. for me personally, Id take a 50-85 f/2 to go with the 24-35 f/2 and those 2 can do 90% of what I need in weddings. id hold on to my 70-200 VR1 and my tamron 15-30VC for the far and UW and these 4 lenses for wedding coverage is enough.

            • Spy Black

              That’s not your Pappy’s Sigma anymore dude. You gotta leave that frame of mind behind. If they build it, they will deliver.

        • 24×36

          If you’re using it at f5 or higher f numbers, it’s probably better than the Nikon. So much for it being “garbage.” It depends on how you’re using it.

          • fanboy fagz

            no

      • Eledeuh

        Hah, fstoppers is probably the last place to get interesting insight on gear.

        Lee Morris and Patrick Hall have done some very cool tongue-in-cheek stuff about gear in the past, but the other writers are completely on the buzz-y side of the business. They don’t provide any useful information, worse actually, wrong/misleading information.

        DXO has lots of limitations but I guess if they’re taken into account you can get some information out of it 😉

      • 24×36

        Stop looking at the “overall score,” because that is nothing but an average, and tells you very little. Their results are perfectly fine, you just have to dig into the details to see where the differences, advantages and disadvantages lie.

        Fact is, the Sigma 24-70 f2.8 is only weak compared to the Nikon 24-70 f2.8 if you’re shooting wide open. Stopped down, it may in fact be better, since it has less CA than the Nikon, with equal (or close to it) sharpness. In fact, the Sigma often has better sharpness across the frame. If you stop down, the Nikon is more of an “optical disaster” than the Sigma is. It all depends how you use the lens.

  • Davis5

    for once they are right, the 18-200 tamron is a lens too cheap to be used on a decent SLR…

  • peter w

    I would think the Sigma 24-35 should not be compared to 24-70 zooms which are all round lenses, but rather to 24 and 35 primes and wide angle zooms like 17-35 F2,8 or 16-35 F4.

    • Eledeuh

      Even comparing them to UWA zooms wouldn’t be super informative. 16mm to 24mm is a huge difference, and makes quite a difference in terms of design too.

      This Sigma really is more a sort of “multiple-primes-in-one” lens and comparison with other families of zooms don’t give a lot of insight.

    • fanboy fagz

      Actually nasim revoewed the sigma 24-35art vs nikons 24 1.4g/ 28 1.8g/ 35 1.8g and it outperformed them all. I dont doubt it. I saw images from it. Open closed its just phenomenal

  • Nikita

    Sigma is doing some nice stuff; makes Nikon look bad.

    • Patrick O’Connor

      I agree. Using a Sigma lens to photograph a Nikon would make it look bad! 😉

      • Nikonanon

        Word

  • MonkeySpanner

    I think everyone is missing a big opportunity by not updating the outdated 17-55/2.8 DX zooms. Nikon’s version is getting close to 15 years old. And tamron and sigma designed there versions back when they were not making super high quality stuff like they are now.
    I think a very high quality 17-55/2.8 would fly off the shelves at $700.

    • Sawyerspadre

      Although if Nikon did it, it’s highly unlikely to be $700.

      • MonkeySpanner

        Oh, yes, I don’t recommend Nikon do it. It would cost well north of a grand.

        • Nikonanon

          16-80mm nikkor?

  • 247th

    I’m not surprised how much more sharp the Sigma 24-35 is than the other lenses, it covers such a short zoom range it SHOULD be sharper at 24-35 than it’s wider/longer competitors. I don’t see why DXO compares this lens to the 24-70 instead of the 17-35 or the 16-35’s which are aimed towards landscapes?

    • Patrick O’Connor

      I don’t really see the 24-35 as a landscape lens although you could use it very well for that purpose. More of an environmental portraiture lens and, for that purpose, appropriately compared to 24-70’s. Of course, having a shorter zoom range it should be better. But then “Life’s not fair, is it? You see I…well I shall never be king. And you…shall never see the light of another day.”

  • To me the 24-35mm is for people shooting 35mm prime. It allows to go wider when really needed.

    I will probably sell my 35 Art for that reason as i have had little use of it and often preferred the 24mm 1.4G as I could crop a bit to reach the 28mm 1.8 or 35mm f2 equivalence…

    I will give a try to this 24-35mm and to the new 20mm 1.4 but I am still a bit skeptical about the AF reliability with sigma.

  • 24×36

    Great. Now make one just like it, with NO “OS” and similar excellent optical quality and all the rings in the same places and turning the same way, only make it 24-70 f2.8. 😀

  • Back to top