DX super-wide zooms

Discussion in 'Nikon' started by Dieter Schaefer, Jun 15, 2018.

  1. I just went through comparing three DX super-wide zooms:

    Tokina 11-16/2.8 (the first version)
    Tokina 11-20/2.8
    Tamron 10-24/3.5-4.5 Di-II VC HLD

    The intent is to replace my wife's 11-16 with something a bit more versatile (which brought the Tamron into the picture because of both the widest range and the presence of vibration reduction). I have not tried the Nikon AF-P DX 10-20/4.5-5.6G VR - which is the other DX super-wide that offers vibration reduction. For what my wife mainly shoots, a wide aperture is not necessary, which automatically makes the Tamron the favorite from the above list. Since of lot of what she shoots is landscapes or cityscapes, somewhat even performance across the frame is desirable - which is why I focused mostly on corner performance in my comparisons (done on a D500 and evaluated at 100%).

    In a direct comparison at the widest zoom setting, one surprise was that the old 11-16 clearly beats the new 11-20 when it comes to corner performance (center sharpness seems to be at par with all three - at least I can't see a significant difference). It's not a big difference but definitely noticeable in a direct comparison. The Tamron looks slightly worse wide open (f/3.5) compared to both the Tokinas. Stopping down to f/4 sharpens up the corners with both Tokinas but does nothing for the Tamron (no suprise really as it is only 1/3 stop down from wide open). At f/5.6, the small advantage of the 11-16 over the 11-20 remains and the Tamron still lags behind but manages to close the gap somewhat when stopped down to f/8.

    At 16mm, the 11-16 looses in the corners against the 11-20 and the Tamron 10-24 at every aperture; the 11-20 clearly beats the Tamron at each aperture.

    At 20mm, in the corners, the 11-20 is clearly ahead of the Tamron at each aperture again.

    The 11-16 performs well at the widest focal length setting and gradually degrades towards the long end. The 11-20 does the opposite, as does the Tamron.

    Overall, the 11-20 is better than the 11-16 except at the widest focal length setting. Replacing the 11-16 with the 11-20 comes with the additional caveat that the 11-20 requires 82mm filters, the other two make do with 77mm. The Tamron never matches the 11-20 but gets close to the 11-16 with the Tokina having the advantage over half of its focal length range (starting with the widest) and the Tamron over the rest.

    I summary, the 11-20 is the overall sharpness and eveness king of the three lenses compared while the Tamron is the most versatile due to the widest range and the presence of VR. And I am nowhere closer in finding out whether I should recommend the 11-20 or the 10-24 to my wife (though I am leaning towards the latter). Her main lens is the Nikon 16-80, with either lens there will be some overlap - which I consider a good thing as it may eliminate some lens changes.

    Anyone wants to chime in on the Tokina 11-20 vs Tamron 10-24 VC selection decision?
    Andrew Garrard and Mary Doo like this.
  2. Good to see your review as I am not familiar with these lenses and some of my photo friends do use and like them but have not offered any detail feedback. I have never used anything other than Nikon for super wide. I wonder how Nikon 10-24 would compare?
  3. Can't tell from personal experience as I have never used one. The much higher price discouraged me from even considering it.
    Mary Doo likes this.
  4. Just watched an excellent review of the Tamron 10-24mm VC HLD. He is all for it, saying it's better than the Nikon version. There are other reviews too and I happen to have watched this one. (Link).
  5. One thing you might want to examine is how the different choices interact with filters. I have the original Tokina 11-16mm f/2.8 and found that even a filter with excellent anti-reflective coating can contribute to flare. Given that the lens hood has to be extremely short, a filter screws into a position where it is highly exposed to light from outside the frame. I returned the protective filter purchased with the 11-16mm and soon found a scratch in the center of the bulbous front element. I have not tried Tokina's 11-20mm, but, from pictures, the front element and hood look similar. The smaller aperture Tamron may have its front element somewhat less exposed.
  6. Good point. Haven't looked at the 11-20 for that aspect but the 10-24 performance in that regard is impressive (two spots in the center of the image, no reduction in contrast).

    Just watched two others - both also quite positive on the 10-24 (and one not so much on the Tokina 11-20 - not faulting its sharpness though). Not putting much trust into what the "Angry Photographer" has to say - even if it says it over and over again (after eventually, maybe, getting on point).

    Dxomark doesn't show good results for the 11-20 - these results clearly are add odds with others (like lenstip) and also don't match my own testing. There are reports about several issues with the 11-20 - like backfocusing or poor performance close up. Dxomark has not yet tested the 10-24 VC.
    Mary Doo likes this.
  7. I wish I could understand dxomark results: the same FX lens on a D810 body is excellent and then shows mediocre performance on a D7000 body - almost the same pixel pitch on both sensors and using only the lens center to create an image yet the performance suffers? Makes no sense - it appears if they would switch the D810 into DX mode, the same would happen - the lens acutance would drop when throwing half the sensor data away; something certainly is amiss at least with how dxomark presents the data. There "how we measure" explanation is no help on that subject either - at least not to me.
    steve_g|2 likes this.
  8. Various review sites normalize resolution by dividing by image height; if that's the case here, a lens with wide coverage (and little quality fall off) but relatively low resolution may well look better on a full frame sensor than a crop, irrespective of the actual ability to resolve detail at the sensor level. I've no idea whether DxO does this (I rarely look at numerical scores, and never between formats).

    I've not checked out this particular review, though. Maybe they had a bad sample?
  9. Ok, i know it is less "wide" but did you also consider the Sigma 12-24mm f/4.0 DG HSM Art ?
    It is less wide bu it is also usable on FX which may compensate for the 2mm difference ( 3mm on DX) ….
    just another option.....
  10. I don't have anything similar to compare it to, but I have no complaints about my Tokina 11-20mm f/2.8. Scaling the image up, I'd say the IQ is as good as the 14-24 Nikkor is on FX.

    I'm glad that Dieter finds the Tokina 11-20 the overall winner. I'd hate to think I'd wasted my money on it over the 11-16 version.

    Maybe weight could be a deciding factor? The Tokina is quite a hefty lens, whereas Tamron usually manage to pare the weight of their lenses to a minimum.
  12. I don't need a wide angle zoom that often but I did buy a Sigma 10-20 4.-5.6 lens recently and on the few occasions I needed to use it,I was satisfied with the results. It does not have VR, which I found I did not need at wide angle shooting outdoors. It is light weight and easy to carry. And it works fully on my Nikon D7xxxx and D3300 bodies. I'm happy with its versatility, price and performance.
  13. At currently $1500 (with $100 rebate) it costs a lot more than I am willing to spend on a lens that won't see that heavy a use. The Sigma weighs in at 2.5 lbs, a lot more than my wife (or even I) is willing to carry; not to mention the vulnerable bulging front element that precludes the easy use of filters.

    I've not considered lenses that have a variable aperture in that range - I don't mind a variable aperture around f/4 (aka f/3.5-f/4.5) but am reluctant to go slower especially when the lens does not offer VR. While still fine for landscape, it limits indoor use too much.

    120g difference between the Tokina 11-20 and the Tamron 10-24 VC - certainly tilts the scale in favor of the Tamron but it's not a deciding factor. The 11-20 currently has a $130 rebate here in the US, which makes it $30 less expensive than the Tamron 10-24; at its regular $600 price tag I think I would pass on the Tokina otherwise.

    That's among the three lenses I have tested, with two copies for the Tamron and one each for the 11-16 and 11-20. I could not confirm
    the lack of close-up sharpness that has been reported for the Tokina 11-20 in one report I've seen.

    Not sure what dxomark does - in their field maps and profiles (which is what I am mostly looking at), they report percentages - so one would assume that different image heights are properly accounted for. Here's an example of what confuses me: Nikon AF-S 18-35 on a D800 compared to the same lens on a D7000 - almost the same MP than the DX crop of the D800 and hence the same pixel pitch; both cameras have an AA filter. On DX, the same lens looks a lot worse - which does not make sense to me: Nikon AF-S NIKKOR 18-35mm f/3.5-4.5G ED on Nikon D800 vs Nikon AF-S NIKKOR 18-35mm f/3.5-4.5G ED on Nikon D7000 | DxOMark
    An acutance map apparently takes viewing distance of an image into account - but at what distance and which image size are considered is not mentioned anywhere on the dxomark website that I could find.

    I think I have finally reached the point where I am going to ignore these MTF-based lens tests - too much sample variation and - especially for super-wide angle lenses - a measurement distance that is all but irrelevant in real life shooting situations. And especially for dxomark, a data representation that appears to have a major flaw and suggests results that quite often do not match what other sites report.

    Fairly certain now that I'll give the Tamron 10-24 VC as the most versatile choice to my wife. The remaining debate now is whether I will return the 11-20 or keep it for myself while eventually getting rid of the 11-16.
  14. Sandy Vongries

    Sandy Vongries Administrator Staff Member

  15. I just got that 10-20 to use with my D7100 - not worrying about the lack of VR switch, since it's expected to be for tripodless travel. Not a lot of pictures taken yet but so far I like what I'm seeing. And it's silent.
    Sandy Vongries likes this.
  16. Wonder what gave you that idea :rolleyes:
    I've handled the AF-P 10-20 and wasn't overly impressed - which is why it didn't make my list. While it does have VR, the lack of a VR switch makes it a non-starter anyway.
    Sandy Vongries likes this.
  17. Yes; the field map suggests the 18-35 holds up reasonably well across the full-frame field (in that there isn't a massive drop-off of performance at the edges). The reported "17 P-MPix" on the D800 roughly correlates to scaling the "8 P-MPix" of the D7000 over the larger frame, with a bit of extra drop-off. I suspect the field map considers "17 P-MPix" to be detailed enough to merit a green colour on the graph, despite being substantially behind what the sensor is capable of, whereas "8 P-MPix" is a bit low by any standards; I'd like DxO to have a better way of graphing this on different bodies, since it's not really distinguishing "good" from "exceptional" other than by very subtle shades of green.

    In other words, the ability to resolve at the D800/D7000 pixel pitch isn't all that good, but the lens maintains this performance well across an FX frame, so the contribution to an image is better on the camera with the larger sensor.

    Guessing, though.
  18. I know it's not on the list and doesn't have VR, but when i need really wide, my 8-16mm Sigma does very well.
    I just leave it on f8 @ 1/30 and Auto ISO and it's sharp edge to edge.
  19. Not on the list because it doesn't take filters.

    Had a bit of a surprise yesterday when I checked the flare and ghosting to the Tokina 11-20 - much improved over the 11-16 to the point where I couldn't get it to show ghosting at all; with the Tamron one or two very small ghosts can be made to appear. In terms of flare resistance, I'd rank the 10-24 Tamron higher than the 11-20 Tokina - but there's not a huge difference.

    I think I'll let my wife decide which one she likes more - if she picks the 11-20, then I will likely keep the 10-24 for myself. If she picks the 10-24, then I might just keep the 11-16 and send the 11-20 back - while the latter is better in some aspects, the difference for me isn't sufficient to warrant the expenditure.
  20. Exactly what happened.
    Mary Doo likes this.

Share This Page