Sigma 17-70/2.8-4-OS [vs] Nikon 16-85/3.5-5.6-VR2

Discussion in 'Nikon' started by hlwimmer, Feb 19, 2011.

  1. I'm searching for a new main/single lens to pair with my D7000. previously the 24-85/2.8-4 served me well on my D100 (and might actually be the answer

    In a lens, I'm looking for versatility without compromising quality. I rarely shoot telephoto past the 85mm mark (DX sensor) and I've rarely wanted a longer lens (and when I do, I might reach for a 70-300), but I have often wanted something wider without too much distortion. I shoot a lot of architecture and "square things", so distortion is a big factor as is a large max aperture for hand-held work and nice subject-ground separation. I also tend to shoot a lot of details, so the macro capabilities of the 24-85 have proven very useful and I've yet to come up against a macro-limitation when using it for close work. Overall clarity is also a concern. I remember shooting with a 10X+ zoom years ago and everything at every length seemed muddy... so i've been drawn to the more modest zooms. I also tend not to change lenses often, so a one-lens solution (with a few others like a 50/1.4 and a tele-zoom for those rare, rare needs ) is a must.

    Concerns in order:
    1: One lens solution:
    ... wide-zooms and primes are purposefully not on the list
    ... I need wider than 24mm (16-18) and something approaching 70-85mm on the long end (50mm is too short)
    2: Distortion
    ... the distortion on the 24-85 isn't bad, but it's terrible on the 18-105
    3: Clarity
    ... my 24-85 seems plenty clear with good contrast
    4: Close focusing
    ... with my 24-85, i've not needed a "real" macro and i shoot enough random close details that a "built-in" macro is handy
    So far, in addition to my existing 24-85, I've paired it down to 2 others -- each with pros and cons (and eliminating another -- the 18-105)

    // Nikon 18-105: (I'm eliminating this one).
    -- I own it (came with the D7000)
    -- distortion + vignetting
    -- 3.5 base aperture
    -- poor built quality
    -- VR1

    // Nikon 25-85/2.8-4macro pros:
    -- I own it
    -- 2.8 base aperture
    -- Fairly clear (by my estimation)... so this might be a good measure against the other lenses
    -- Macro capabilities
    -- Nice build quality (by my estimation)... so this might be a good measure against the other lenses
    -- 24mm is limiting on the wide side
    -- Lack of VR or OS (image stabilization)

    // Nikon 16-85/3.5-5.6
    -- Wide end flexibility
    -- VR2
    -- No macro
    -- 3.5 base aperture
    -- Cost (higher)

    // Sigma 17-70/2.8-4-OS
    -- Cost (lower)
    -- Macro
    -- 2.8 base aperture
    -- OS
    -- 15mm less on the long end
    -- Not a nikon-brand lens (does this matter?)

    It really seems the decision is between the Sigma 17-70 and the Nikon 16-85... and the Sigma seems to be pulling ahead on paper. Your thoughts appreciated (and thanks for your patience and wisdom).
  2. The 16-85 does have slightly less distortion than most kit lenses, but $600+ is too expensive for what it is. I'd go with
    the Sigma. It's just as good, with better close focus and a more reasonable price tag.
  3. i would say a big con of the 16-85 is the 5.6 aperture on the long end. there are P&S cameras which are f/2.5 at 112mm! in real-world terms, this effectively limits the 16-85 to daylight use in good light. generally, i would opt for constant 2.8 across the range, so i would also consider the tamron 17-50 and sigma 17-50. to me this is more important than the longer overall range, but YMMV.
    i'm afraid, though, there are no easy solutions; any lens in this field will represent a compromise on some level. i assume your main reason for wanting to replace the 24-85 is the wider zoom range. but, and this must, be pointed out, any lens you get is going to leave you with three zooms which overlap focal lengths considerably. so i'm not sure i wouldn't just keep the 24-85, sell the 18-105, and buy an UWA like the 10-20 or 10-24, and maybe the 85/1.8.
  4. For the first time I agree with Eric. Keep your current zoom for what you find it excells at, and get another zoom to go alongside it that achieves different purposes. Part of the problem is that any of the generic zooms in the range you are looking at, especially with VR/OS, will likely be poorer performers than the one that you already have.
    I would add the Sigma 12-24 (also good for full frame) and the Nikon 18-35mm f3.5-4.5 D ED IF to the list. For those rare occasions when you absolutely need VR you could get that 50/1.4 instead or add a used Nikon 18-55 VR for less than $100 USD.
  5. Thanks for the early/quick insight.
    Eric: Good point about the 16-85's aperture -- thus the interest in the Sigma (with the 2.8-4.0). FWIW, I've not been disappointed with the 25-85's aperture range (2.8-4.0), so the Sigma might be the best bet -- and with the OS would still be better than the 24-85 (and let me keep the "macro" function)... I'd just compromise the 15mm on the long end.
    The main reason for replacing the 24-85 is indeed the want for a wider zoom in a more "one-lens-solution" package, however 4 lenses is simply too many for me to carry around and realistically swap/use. Honestly, the need for a wide angle less than 16 or 17 is minimal for me. It seems with the 17-70 and a tele-zoom (70-300), i might have all of my needs covered in 2 lenses.
    If someone made a low-distortion 2.8-constant 17/18-70/85 at any price, I'd be game... but it seems with the 2.8s i'm either giving up the wide end (Nikon 24-70/2.8) or the long end (sigma 17-50/2.8).
    Andrew: Fortunately, the cost of the 16-85 isn't as much of an issue as the compromises I might make with it (lack of "macro" and higher base aperture). The only thing it seems I'm gaining with the Nikon lens is the name, 1mm on the wide side and 15mm on the long side...
  6. It seems with the 17-70 and a tele-zoom (70-300), i might have all of my needs covered in 2 lenses.
    if you dont need to go wider than 17 or shoot in extreme low-light, that might indeed be the case. i'd still want at least one sub-2.8 prime for available light, but i dont see the need for three overlapping zooms, especially if none of them is a constant 2.8.
    If someone made a low-distortion 2.8-constant 17/18-70/85 at any price, I'd be game...
    well, there's the rub,isn't it? many years ago, there was a tamron 28-105/2.8 which produced inconsistent results. but a similar lens for FX, and maybe a DX variant which started at 18, with OS/VR/VC, would have immediate usefulness.
  7. >>If someone made a low-distortion 2.8-constant 17/18-70/85 at any price, I'd be game...
    If it had VR or it's equivalent, it would be the size of a coffee can. I like the 16-85, which I use mostly at it's wide setting. 24mm was my favorite focal length on 35mm. For many, the f5.6 @ 85 is marginally worthless, but for the photography that I like, it works fine.
    I also generally prefer Nikon glass, as their lenses built their reputation more so than their camera bodies. (Yes, a 3rd party lens or two may reside in my "go to" bag.)
  8. @hunter
    One of the reasons you see such little distortion with the 24-85 is that is meant to fill a 35mm frame. Shooting on DX you are using the "best" and least distorting part of the glass - the center. If you shoot it on film or FX you will likely notice the distortion around the edges a bit more.
    Unfortunately, to get anything wider than 24mm in a non-DX Zoom, you will be limiting yourself to lenses that cost several thousands of dollars and only go to 35mm on the long end.
    The answer: Compromise.
    Any 16/17/18 - 70/85mm DX lens is going to have distortion. The question is: how much distortion can you handle, and how easy is the distortion to fix in post?
    How much distortion you can handle is up to you, but nearly all (simple) lens distortion is now very easily fixed in post via Nikon and/or Adobe software.
    If you are looking for a one-lens solution, get the 16-85mm f/3.5-5.6 The f/5.6 on the long end should not worry you if you are shooting mainly landscapes and architecture. For hand holding, it may become an issue, but the VR is quite good and should allow you to shoot at 1/30 - 1/60 sec @ 85mm with little to no hand shake noticeable. I don't own this lens, so if I'm wrong on this, someone with more experience with this lens should correct me.
    If you don't mind having two lenses, keep your 24-85 and get the 12-24 f/4 or 10-24 f/3.5-4.5.
    Hope this helps
  9. Thanks, all: I definitely appreciate the wisdom. Some interesting points ala "coffee can" (of a 17-85/2.8VR) and "sweet spot" (of the 24-85).
    Presently, I'm leaning towards the Sigma for the maco functionality and wider aperture -- it seems to spec out better than my current 24-85 in tests on photozone. I'd give up 15mm on the long end, but again, "compromise" seems to be the word here.
    Am I nuts? (Sigma over Nikon)
    (I'm also intrigued by Sigma's 50/1.4, but that's another post... ;) ...)
  10. No, you're not nuts. I've used both lenses (we're talking about the Sigma 17-70 and the Nikon 16-85, right?) and they're equally good. I have 13x19 prints of shots I took with the previous (non-VR) version of the Sigma and they came out very nicely. I did eventually sell that lens because it overlapped another lens I have but I tried the new version and it's just as good, and I'd be seriously thinking about buying it myself if I weren't also seriously thinking about moving to FX whenever Nikon gets around to releasing the D700 successor.
  11. The sigma is a very interesting lens, and the wider aperture esp. (in my view) makes it so. The 16-85VR is a seriously good lens, pricey for what it is, but it's (at least mine is) usable wide open. Versatile, sharp, contrasty and a good tool. But personally, as much I like having the 16-85, it also begs for one or more fast lenses alongside it. The aperture is limiting. The Sigma might lessen that 'problem' a bit. So, nothing nuts about it.
    However, I would personally not dismiss the 18-105VR that easily. Distortion is easy to fix, and VR generation 1 versus 2, I wonder in real world usage whether it really makes that much of a difference. The build quality might be a concern, but since you indicate not wanting to change lenses all that much, you won't see it plastic mount all that much.... I'd get started with this lens and see how it works out. It's a lot of lens for little money, and it might surprise you. Spending money on a better lens (or lenses) is an option that always stays.
  12. I would say that unless you absolutely need the speed, go with the Nikon. The reason the Nikon is so expensive is because Nikon actually has a QC department, and the build quality of even their cheapest lenses far exceeds that of any Sigma lens. Likewise the Nikon auotfocus is likely to be a whole lot more reliable, especially under low light… and vicious (probably unfounded) rumors abound about Sigma's image stabilization (OS/VC/VR) using more juice than Nikon's.
    I got my 30 back from Sigma, and just now tried to clean some schmutz off of the barrel with my fingernail. Thing is they replaced the chipped barrel (with the new finish) with one utilizing the old velvet finish. Needless to say it peeled right off with my fingernail. The AF is a /BIT/ improved, but hunts way more under low light than the two previous copies (and way more than any Nikon lens I've used).
  13. I am using the Sigma 17-70 2.8-4.5 on my D7000. It is the previous iteration of the lens mentioned above. I had bought the lens to use on a D60 which my wife appropriated and she had a different preferred lens so the Sigm sat in drawer for a while. I did not find the kit lens acceptable so dug out the Sigma and have used it a lot since. I have a 10-20 Sigma for wide stuff but the 17-70 is actually a pretty good lens. I think the 17-55 Nikkor is undoubtedly better but I like the 70mm long end and the fact that it goes no lower than 4.5 on the long end. 5.6 is pretty restrictive and keeps me from getting the 16-85. I think you will not be disappointed with the Sigma.
  14. the build quality of even their cheapest lenses far exceeds that of any Sigma lens.
    Likewise the Nikon auotfocus is likely to be a whole lot more reliable, especially under low light…​
    alex, your experiences with sigma might seem frustrating, but you seem to be judging every lens they ever made by a poor experience with one lens. i'm not personally in the habit of running a fingernail down the edge of a lens to see if i can get the paint to crack, but i have several sigma EX lenses, the build quality is better than tamron as well as low-end nikkors, and the AF on my 50-150, 30/1.4, 50/1.4 and 15/2.8 fish are all pretty good.
    AF also depends on what camera you're using. i've found the AF on the sigma 30 to be pretty accurate on the d300 and d300s bodies. and, the d300s is actually faster to focus with the 50-150 in available light than the D3s with 70-200 II. why? because the D300s has an AF assist light; the D3s has none. once the d3s locks focus its a speed demon, but i've missed shots with that i would have gotten with the d300s. of course, i rarely go above ISO 2000 with the d300s, but still... the point i'm saying is you can't have everything.
  15. The paint isn't cracking, it's peeling off. I just threw away a chunk of velvet. This is /after/ Sigma said they fixed both issues — altho utilising the older finish which is known to be problematic smacks of not really making much of an effort when the lens I sent in had the 'improved' finish. FWIW, my fingernails aren't made of diamonds (or even very long at this point). This is my third copy, after Sigma shipped it back to me with a smaller box and less protection than I used. I sent it to Sigma using just the Amazon stuff as-is — and Amazon catches a lot of grief for how poorly they pack their lenses.
    Getting AF to work reliably can be tricky because there are so many variables. Getting paint to stick to metal /can/ be tricky, but is generally well understood (esp. when you're not building it in the Eurozone or California). And packing these things properly (f/e.x. with more than just one air blister) is sure as hell not rocket science. How many samples do I go through before I'm allowed to point out that none of the Nikon lenses I've used have been this poorly built or unreliable? Hell, this third copy fits pretty damn tight. It's as if the tolerances for the bayonet mount are *X&$#**X&$#**X&$#**X&$#*.

    When I rent the 12-24, I rent it from Gassers. I was one of the first people to rent it, and saw it in pristine condition. Last time I rented it it looked well used, but it still hunts less (but after a year or so of hard rental use it looks slightly worse than my brand new Sigma). I leave the AF assist lamp on and mainly use the center focus point. Even with the lamp this Sigma's AF is super jittery. At one point it did its jittery thing for > 1 second (AF-S not AF-C). Sure there's a bit of brassing on my 28, and yeah there's some paint worn on my 58… but the 58 has been my primary lens for /two years/ and it looks and works better than my Sigma… this last copy I've had for less than one week (been in my hands since Wednesday). Sigma is aware of the model of camera I'm using, and should be aware of the firmware in that camera as they've got a CD of NEFs from me demonstrating the problems.
    No, I don't think I'm asking too much for the finish to stay on the lens barrel and the autofocus to work consistently. In fact if you look at one of the more recent Flickr threads, you'll see one or two other people complaining about not getting as many keepers from the 30/1.4 as they do with the 35/1.8. If you go back further, you'll find more of the same. Nikon can make those things happen for $100 (and, in fact, the non-D 50/1.8 that I'm borrowing looks /much/ better and works more reliably than my Sigma that cost nearly five times as much). Nikon's /cheapest/ lens works better and is built better than Sigma's top of the line stuff. If Sigma were marketing this as an entry level lens at $100-$200 that'd be one thing. Instead, they've branded it "EX" and are asking a premium price. Think about it for a sec. I'm trying to hold a $450 top-of-the-line Sigma lens to the same level of quality I'd expect from a $100 cheapie Nikon lens. Should I even mention the infamous Zen finish or the problems that had both with Sigma QC and Sigma USA service?
    My opinion stands: Sigma does not build pro grade lenses. They design and build nice lens elements, but everything else is *X&$#**X&$#**X&$#**X&$#*. If you want something that you can throw in your bag and expect to just work, get the Nikon. The reason that a constant aperture Sigma is priced competitively with a slower, variable aperture Nikon is because Sigma cuts corners plain and simple.
  16. Thanks again for the responses and insight.
    I /am/ leaning towards the 17-70/2.8-4 Sigma as a general "walk-around" lens because of the speed, range, macro and OS. Nikon seems to have lenses that fall into one or two of the four, but not even three (i.e.: the 16-85 has the range + VR, but not the speed nor the macro whereas the 24-85 has the speed and the macro, but not the range or VR)... so, I figure I'm compromising less with the Sigma. Hopefully Alex's Sigma-Karma doesn't wear off ... ;)
    Alex -- you seem to be local (i.e.: SF-based, like me). While I work about a block from Gasser's, I gave up on them long ago on everything. However, renting few lenses does make sense... especially if a shop carries Sigma in it's rental line.
    Locally, I can think of Calumet + Gassers for lenses and Photo-supply for general (lighting) rental... then K+S all the way down in Palo Alto. What other spots are there? (SF seems to have terrible resources considering it's size.)
    [[As an aisde, I'll be looking for a digiback for a 500-series Hassy in the fall (CFV39 or CFV50). K+S used to rent it, but they've become spotty with what they cover. Ideas? I s'pose I should just start calling around...]]
  17. i received the sigma 17-70/2.8-4 (and the 10-20/3.5-EX) and preliminary use seems to indicate that it's at least as good (sharpness, distortion, etc.) as the nikon with the same(ish) macro abilities... so, other than the 15mm on the long end, i'm not missing much (save future re-sale value) -- and gaining: OS and a wider field of view.
    i might even go so far as to say it seems charper, but i'm using it on the newer D7000 (ratherthan the old D100), so there's lots of differences technologically body-wise.
    the 10-20 is likewise a nice lens and perfect for the architectural work i do.
    it's also nice that photoshop has the distortion profiles for both lenses... it didn't for some of the nikons, oddly enough.
    so... in the end: thanks for the insight and wisdom. i'm happy with the purchase(s) so far and will report back if there's something significant to add otherwise.
    eventually, i plan on picking up their 50/1.4 and a longer (70-200/2.8, perhaps)... and will report back.
  18. Hunter: I've only dealt with their still rental department, and had good luck with them. I don't get warm fuzzies from the other guys tho. I've been working from home (across town) more now, but it was super convenient when I was working a couple blocks away. Unfortunately they do not stock off-brand lenses for rent. is also fairly local (and delivers to a few places in the city), and atrocious to deal with. I tried to rent a Sigma 30/1.4 from them before buying and despite two confirmed reservations, they managed to cock it up both times and finally (despite explaining that I wanted the Sigma specifically because I was considering its purchase), they tried to sub an optically inferior Nikon that was of little interest to me (and not particularly close in focal length). But, yes, renting (online or in person) is a good idea before buying. There are a couple of online shops I'd consider, but the conclusion I've come to is that San Francisco (and the Bay Area in general) is not New York. We don't have the food or the photo shops.
    'Course right now I've about had it with Sigma's service. Apparently all of the test shots I sent were for naught because they have /NO/ idea what to do with a CD of NEF files (let alone one NEF). Un freaking believable.
  19. Thanks, Alex: Fortunately, I must've gotten a good batch (or my standards are low) from Amazon. When I purchase the 50/1.4, i might order three and return two... ;)
  20. Yeah, Amazon is great to deal with (despite my misgivings about patronizing them), exchanges aren't a problem at all. seems to be more legit than BorrowLenses (both more organized and inspecting their equipment properly), altho you don't have the option of picking it up in San Carlos. They took Sigma (and Sony) to task for their atrocious service and reliability. Me I'll be saving up for the Nikon 35/1.4, with the Sigma bubble wrapped in a closet somewhere… lest the paint start coming off again.
    Seriously Sigma's New York center owned up to having copies of photoshop but still had NO IDEA how to open an NEF. I'm going to be ranting about that one for years to come.

Share This Page