I recently picked up Sigma's 17-50mm F2.8 EX DC OS HSM lens (from herein referred to as Sigma 17-50mm) as I had a need for fast lens covering the standard zoom range. With a few weddings ahead of me I thought this would be a good lens to have in my kit, and it fit my budget, coming in at right around $500 USD.
After much online research comparing the Sigma to Tamron's similar offering, I decided to go with the Sigma as I had used one of their lenses in the past and quite enjoyed it. Nonetheless, before the lens showed up, I'll admit I wondered what quality of glass I would be getting at such a reasonable price.
Before I get into the specifics of the lens, I'm going to put this post in a little context. I'm not a lens reviewer. I know basically what to look for when assessing a lens, but I'm not set up for or familiar with all of the specific tests that more knowledgeable lens testers are in this regard. That being said, this review will be more basic than others, essentially, just a guy shooting with a lens and looking for some of the more obvious flaws one might pick up.
After taking the lens, its padded case, front and rear covers and hood out of the box, I gave it a once over to make sure visually everything looked intact. All screws were in place, and a quick double check with a jeweler's screwdriver confirmed all showing hardware was tight. No audible sounds were heard when giving the lens a little shake.
The lens' zoom operated smoothly, though much tighter feeling then my other lenses. When extended, the lens measures just over 3.5", with the front element telescoping out from the barrel. I was briefly confused by the zoom ring operation, which is reversed from the Nikkor lenses I'm familiar with. Fully extended, I could feel no play when gently rocking the front element in the barrel. A quick switch of the AF/M switch proved that the manual focus ring also turned smoothly.
Despite being largely plastic on the exterior, at 565g, it's not a light lens, but certainly nothing so heavy that I wouldn't keep it mounted as a walk-around lens. It does have a metal bayonet, which is something I always look for when purchasing a new lens -- learned this lesson when a plastic mount on a kit lens broke, resulting in a bent aperture arm, and a repair bill from Nikon USA!
After the visual inspection, I felt pretty good about my purchase. Of course, the real tests hadn't even begun.
First off, the Sigma 17-50mm is specific to APS-C sized sensors, so its angle of view is similar to that of a 25-75mm lens on a 35mm film camera. For my needs, it's plenty wide enough to capture the view of say a wedding ceremony with guests, and long enough to get in a bit tighter on the scene. The 50mm end of it also means it should perform well as a portrait lens, given that it's functioning like a 75mm.
As for autofocus, this lens is both fast and quiet. I didn't record audio, but compared to other lenses I've used, the Sigma is barely audible when focusing and it's quite quick at locking on to a subject. In low-light, it can tend to hunt a bit, but overall it's a very good performer in this regard.
I figured I'd like to get a sense of how much distortion the lens had and how sharp it was, so I set up shooting a wall made of vertically oriented rough cut lumber. I made a number of shots at the 17mm, 27mm and 50mm focal lengths, and with the aperture set at f2.8, f11, and f22. At f/2.8, the lens shows some barrel distortion and slight vignetting, but nothing so serious at the wide angle that I'd give it bad marks. In fact, once imported into Lightroom 3 and hit with the lens correction settings for the lens, the distortion and vignetting were hardly noticeable.
As for sharpness, I expected slightly soft results with the lens wide open, but was pleasantly surprised. The center of the image was very sharp, and while this definitely fell off along the edges, it was still well within my acceptable limits.
At the 27mm and 50mm focal length at f/2.8, neither distortion nor vignetting were noticeable to any extent worth mention.
Stopping the lens down to f/11 showed that the lens was definitely a wise purchase. At all three focal lengths, both the centers and the edges remained quite sharp, with distortion occurring again only at the 17mm focal length.
At f/22, the Sigma 17-50mm does seem to lose a bit of resolution, but this is also to be expected to some extent when a lens is stopped all the way down to its smallest aperture. Nonetheless, I was still pleased with the results, though I'm not going to post another image of the wall as it's a little tough to discern this loss of resolution in this subject. You'll just have to take my word for it.
I took a few shots of a church steeple against the sky to test for chromatic aberration, and the Sigma 17-50mm did show a bit of this. The image below is a 100% crop taken from a photograph made at 17mm focal length, f/2.8 1/160th ISO 100. It's not very bad at all, but there's certainly a bit of CA occurring along the lightning rod and the roof lines. Still, nothing I'd give bad marks for, and not anything that couldn't be easily handled in post.
Of course, there are two things I was very excited about testing with this lens: it's bokeh, and its optical stabilizer, which is claimed to allow you to slow the shutter an additional four stops. As for bokeh, I love it. Again, I'm not a professional in terms of lens testing, but to my eye, this lens produces creamy backgrounds when opened up.
Now, I'll admit, I suffer from the shakes when shooting. So much so that when I get down to an 80th of a second, it's questionable whether I'll get even a soft shot. So the OS functionality of this lens was something I was looking forward to, and I wasn't disappointed. I started out a 1/5th, just because I'm an optimist. The results weren't so hot, but had I been able to hand-hold at that speed and get something usable, I'd have probably passed out from shock, dropped my camera and broke the lens anyway, so it's probably a good thing. Shockingly though, at 1/15th of a second, I was actually able to consistently get half-way decent photographs. It's very rare that I would actually try this outside of a test, but it's nice to know I have a bit of wiggle room when working with this lens.
Overall, I'd say this is a winner of a lens. Now, I'll go back to the beginning where I said I wasn't a professional lens reviewer. And I'm still not, but for my purposes, I can safely say I've tested this lens to the best of my ability, and it'll do the job.
All of our Sigma articles
We have several resources specific to Sigma and suggest you have a look at the Sigma index.
You have a question regarding your Sigma lens? No problem - just ask in our Sigma forum.
More articles that might interest you