Review – Sigma 10-20mm f/3.5 EX DC HSM for Nikon

I had an unfortunate mishap with my wide-angle lens, so I had to replace it. I chose to try the new Sigma 10-20mm f/3.5 wide-angle lens rather than choose the Tokina 12-24 (sharp but still CA problems, odd AF/MF switching), the new Nikon 10-24 (initial reviews say that at 10mm the IQ is so bad it’s not worth it compared to the 12-24), or the Nikon 12-24mm (expensive). I used to have a Tokina 11-16/2.8 but it was a bit soft and the focal range is too narrow for me, so I didn’t consider that lens this time. I thought the new fixed maximum aperture design of the Sigma (the older version is f/4-5.6) might lend more Image Quality, so I tried it.

I’m no official lens tester, but since I seem to have a fairly early copy (B&H and Amazon showed it as pre-order status, got mine from Adorama) I thought I would take some sample images and share them. I shot some images of a fence at different focal lengths and different apertures. I used my D300 and used a tripod for each shot. I shot using 14-bit RAW then loaded the files in Photoshop CS4 using Camera Raw 5.4 with all settings zeroed out (no adjustment to RAW data).

Sorry the images aren’t perfectly straight. I decided not to re-do them because this set shows the distortion of the lens. The images are a bit under-exposed too – sorry.

People buy lenses like this for the widest focal length, so I tested that first.

10mm f3.5 center

10mm f/3.5 100% crop, center of image

10mm f3.5 corner

10mm f/3.5 100% crop, top right corner

10mm f5.6 center

10mm f/5.6 100% crop, center of image

10mm f5.6 corner

10mm f/5.6 100% crop, top right corner

10mm f11 center

10mm f/11 100% crop, center of image

10mm f11 corner

10mm f/11, 100% crop, top right of image

10mm f22 center

10mm f/22 100% crop, center of image

10mm f22 corner

10mm f/22 100% crop, top right of image

I’d say at 10mm the lens isn’t super-sharp, and there are some noticeable CA areas that need to be fixed in post. It’s really wide, though. I’d rate it at 6/10 at the 10mm length, since the issues can mostly be fixed in post.

I’ll show 14mm next, since that was my widest focal length with my old lens.

14mm f3.5 center

14mm f/3.5 100% crop, center of image

14mm f3.5 corner

14mm f/3.5 100% crop, top right corner of image

14mm f5.6 center

14mm f/5.6 100% crop, center of image

14mm f5.6 corner

14mm f/5.6 100% crop, top right of image

14mm f11 center

14mm f/11 100% crop, center of image

14mm f11 corner

14mm f/11 100% crop, top right corner of image

14mm f22 center

14mm f/22 100% crop, center of image

14mm f22 corner

14mm f/22 100% crop, top right corner of image

So, IQ seems to be quite noticeably improved over the 10mm focal length. I rate the lens at 14mm about 7.5/10. The 10mm-12mm focal range seems a bit softer than the rest of the focal range of the lens. Aperture-wise, the lens is a bit soft below f/5.6 and there’s a bit of diffraction at f/22, so I will probably be using this lens from f/5.6 – f/16. For landscapes, I’ll probably use something in the 12-14mm range, at f/11 or f/16. I don’t think I would use the 10mm length unless I thought I couldn’t make the shot without it and was willing to spend the extra time in post.

Distortion is something that can be fixed in Photoshop, as are the chromatic aberration issues, so very nice images are possible with this lens as long as a bit of post processing is involved. Here’s a shot that shows the CA issues:

12mm f8 chromatic aberration

12mm f/8 200% crop – There is some purple fringe on the top of the hinge, and some green/yellow color on the bottom of the hinge. It’s there, and it’s a problem, but it’s fixable.

One interesting note with this lens – it focuses extremely close! I thought my Nikon 14-24 was good at getting close, but this lens is ridiculous. I was able to get a shot at 10mm with the lens (hood) about 1.5 inches from the fence! That’s probably 1 inch closer than the Nikon would do.

10mm f5.6 closeup

Overall, I give this lens 7 out of 10. The focal range is excellent for ultra-wide on a crop-sensor camera, and the fixed aperture is a nice improvement. The lens hood mounts easily. The focus ring is stiff, but I love that it will auto-focus while you hold the ring (like Nikon AF-S lenses) and I probably won’t do much manual focus with this lens anyway. The zoom ring is smooth and well-damped – not too loose, not too tight.


24 comments on “Review – Sigma 10-20mm f/3.5 EX DC HSM for Nikon

  1. ken says:

    FWIW, to get the most out of the old 10-20, I had to run it at f/13 and use a tripod. You’ll never get pixel peeping sharpness for the simple fact that you’re compressing a lot of data into 10mpixels. E.g., a 10′ tree in the distance in a landscape picture gets mapped down to 6pixels. You get the idea…A tripod is also a requirement to get decent photos out of it because sunrise/sunset photos require long exposures at f/13…

  2. F says:

    Good review! Is the lens big? Would be nice if u can post some pics of the lens, have just seen one image of it.

  3. Shane Laake says:

    Thanks for the quick review (the only one of any kind I could find). I sold my 10-20/4-5.6 on Craigslist to either get the Nikon (read similar accounts of its 10mm performance) or this Sigma. All in all I was happy with my old Sigma (seemed very good at f11 and good enough everywhere). Anyway, would it be possible to post some other images (anything else)? They could be straight jpegs. I’d really appreciate it.Shane Laake

  4. Lance says:

    These have got to be the most poorly done “review” photos that I have ever seen posted anywhere. They are so horribly underexposed that they look simply awful. A 5th grader could take better photos than this.If this is the best the author can do with this lens, then it certainly must be junk.Lance

  5. Kevin says:

    Oh Wise and Powerful ReviewMaster, I beg for your forgiveness. I am your servant, please teach me. Certainly, amateur photographers have no business performing services for the photographic community, I don’t know what I was thinking. I obviously have no idea what a good review should look like, because of all the hundreds of reviews I’ve looked at, I have yet to find a review that used high-quality artistic images for review comparisons, so I must not know where to go to find expert photography gear reviews with mind-blowing artistic sample photos. Could you perhaps find a moment of your valuable time to enlighten us with your perfect reviews? I humbly beg the forgiveness of the photographic community for taking time out of my vacation to post such a trashy review.

  6. Kevin says:

    Thanks for the tips! They were really helpful, as I was on vacation and really wanted to get some good wide-angle shots.

  7. Kevin says:

    I’ll see what I can do. Product photography is not my fortè.

  8. Kevin says:

    Thanks Shane. I’ll see what I can do. So far I only have private vacation images, but I’ll probably have some images to share in the near future.

  9. Shane Laake says:

    No worries, I found a few on Flickr and got the lens for myself last Friday (and love it).

  10. martin says:

    Thanks Kevin for the review.Yeap, it could have been better, but I am not sure I would be able to do a better review than this.Thanks for all the data that you troubled to share with us. It helped me to get closer to a purchase decision!

  11. eNo says:

    Interesting results. Looks like it will be another popular 10-xx by Sigma. FWIW, I used to own the f/4-5.6 version, and I don’t find my new copy of the Nikkor 10-24 any worse than the Sigma at 10mm. In fact, I think I get more consistent results with it, and I think it’s because the Nikkor is a more consistent focuser. I have some side-by-sides at:

  12. Luka says:

    Thanks very much for posting this review. I can see from the graphs on SLRGEAR that this lens has a major CA problem. I am wondering how well it clears up in Capture NX or Adobe Raw.Could you please send me a couple (or at least one) raw .nef file for experimentation. I would be very grateful as I am thinking of purchasing this lens. My email is luka.strnisa “at” Thanks

  13. Kevin says:

    I think SLRGEAR must have a bad copy, I haven’t seen a “major CA problem”. No lens is perfect, of course. but I think I got a good copy.

  14. Luka Strnisa says:

    So how about that nef sample, it would be really helpful.

  15. Kevin says:

    I’m beginning to think you’re serious. Yikes! I wouldn’t hand out raw files any more than I would hand over my bank account number.What you can do is take your camera body to the camera store and test the lens on your camera body. Or, take a memory card and see if they will let you test the lens on a test body with your memory card in it. Given that not many stores carry Sigma lenses in stock, you might order it from a place with a good return policy, like Amazon. That’s what I usually do.

  16. Hi Kevin,I wouldn’t say that our sample of the 10-20mm f/3.5 had a major CA problem, it was just statistically a bit more prominent than we noted on the f/4-5.6 version. Unsurprising, given the larger front element and aperture, and the price you pay to get f/3.5 at 20mm rather than f/5.6.The other thing to consider is that you’re shooting with a D300, which (for JPEGs) has automatic chromatic aberration reduction. This will reduce the amount of CA you’re seeing; I’m not sure if your shots above are converted NEFs, which don’t include CA reduction, or straight JPGs.In the end, you’re either happy with the results and keep the lens, or unhappy and return it; I’m glad you got a good copy. I’ve owned the 10-20mm f/4-5.6 for some years now and have found it really useful.Regards,- Andrew

  17. Kevin says:

    Andrew, thanks for the comment. I’m surprised that the newer version would have more CA than the older version, simply due to a stronger optical formula (fixed max aperture rather than variable). Simply having a slightly larger front element and larger max aperture don’t translate into having more CA – if that were true, my Nikon 14-24 f/2.8 would be incredibly bad. Just my logic though, I’m not a pro lens reviewer. :)I always shoot 14-bit RAW for my landscapes, process the NEF files in Lightroom 2 and Photoshop CS4, often using Nik Software plugins. I just haven’t noticed any CA issues with this lens. I’m just an amateur though, maybe I’m not looking as closely or something.

  18. Pat says:

    Kevin – Thanks for the review. It is now October 15 and there is still not an official test of this lens. (Other than one that I can find in the UK and it doesn’t include images.) I always like the opinions of owners/users of a particular item. Your images might be a bit ‘underexposed’ but they are also ‘real’. Regardless of the quality the posted images, I think that an owner’s overall opinion and likes and dislikes are as important (if not more) than than tests that are a result of a particular measuring instrument. It very much reminds me of all the stereo equipment reviews in the 70s and 80s. Back then, (and still today to a certain extent) products either succeeded or failed based on certain numerical results, regardless of how that amp, speaker, etc… actually sounded to the human ear. Anyway, I appreciate your effort. I am considering this lens for real estate photography. Have you done any interior shots with it? If so what is your opinion?

  19. Kevin says:

    Sorry, I haven’t done any interior shots with it. I am really enjoying it, though.

  20. Simon says:

    Thanks for the review, Kevin. I own the Nikon 10-24 and 12-24 but I am very curious about this Sigma lens.The reason why the Nikon gets slammed is that @ 10mm f3.5 *extreme corners* are very soft. It looks very similar to your sigma corner tests above @ 10mm. As you stop down to f5.6/f8, things improve considerably. (It’s the same behaviour with the new Tamron 10-24 f3.5-4.5, too)cheers,Simon

  21. Marc says:

    Thanks for the review Kevin! I bought the Sigma 10-20 F3.5 four weeks ago and returned my copy of the 10-20 4-5.6 definitely. The 4-5.6 was returned twice because of unsharp (focussing) results on my Sony A700/A350 and new A550 DSLR’s. After paying an additional 68 euro’s I had the F3.5 model. I prefer fixed F-figures.I have to say that I am very pleased with the 10-20 F3.5 lens! The CA (corners) you mention is a little worse compared to the 4-5.6 version but only truly visible with tree branches against a much lighter sky. With interior photography (no harsh contrasts) CA was not a problem. At 10 mm the 3.5 version seems sharper over the entire frame at large apertures from 3.5 up to 8. Sharpness between F8 and F11 is about the same for both lenses, sharp but not ‘tack’ sharp like fixed focus primes. At the long end (20mm) the 10-20 F3.5 is definitely sharper (less soft) compared to my F4-5.6 copy. Color rendition of the lenses is the same. The F3.5 model is significantly larger than the 4-5.6 version and the filter size is a huge 82mm. A professional grade polarizer set me back 115 Euro… ;-(.The zoom and focussing ring operate fantastic! Nice and smooth and without play. I already shot a lot of shop/hotel and other interiors/exteriors with the 10-20 F3.5 lens. The distortions/light fall off are pretty well controlled and easy to correct in software for best results.The lens is an ideal match for the Sony A550 for this purpose. Using the live view with screen tilted up on a tripod and wire remote. Manual focus check LV with grid and 100% sensor coverage for composition. Normal Live view to fine tune lighting and white balance on screen if not shot in RAW. Light interiors with dark details (see link below) are shot in Auto HDR (JPG only). The A550 shoots 2 frames rapidly (one over exposed and one under exposed)and combines them to obtain the highest possible dynamic range.The Sigma 10-20 F3.5 is also a good available light perfomer, also wide open thru the entire zoom range. With the 3.5 you win about 0.5 to 1.5 stop, the difference between choosing 800 or 1600 ISO – more or less noise in the final photo. The in body ‘steady shot’ of the Sony A550 makes it possible to get very good results at night with ISO 1600, F3.5 up to 1/5 second hand held!See some Sigma 10-20 F2.3 interior photo’s at

  22. André says:

    I also bought this lens and it is excellent. But… I have one problem with it. I want to know if you also experience this or that is is only so with my version of it. The lens hood does mount very weakly. That means it is not really tightly adjusted to the lens which results in the lens hood coming off to easily. Or when shooting photos the inside of the lens hood will be seen in the pictures. It only needs a mm or so to unscrew to give this problem. I would like to hear your experience with the lens hood. I’m having an Sony alpha mount version but that should make not any difference.

  23. Kevin says:

    Mine is not very tight, surely, but I don’t think it’s quite as loose as you describe.

  24. Christine says:

    It really annoys me when someone insults another in the way Lance did. First, someone is doing their best to give an opinion in order to provide helpful info for others. Secondly, an amateur will not boggle less experienced minds with gobbledegook that lesser mortals cannot make head or tail of. Thirdly, I HATE rudeness.There, got that off my chest. Here is an in-depth review on the Sigma 10-20mm 3.5 here is a shorter review, with some positive comments. bought the lens about 2 weeks ago. I am blown away by the width of scene I can now get into a photo – so much so, I put my 18-55mm on the other day and felt total frustration at having to move back to get a scene into a shot.That said, I have had less useable photos – or, should I say, less photos I am impressed with quality-wise. I honestly believe that this is down to me having yet to get used to the lens. I think I am a bit trigger-happy in my excitement! Hence, I am still reading reviews and searching out other people´s views and hints.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s