Some of my photos on covers:
A Review of the Sigma APO 150-500mm F5-6.3 DG OS HSM, and a comparison to the venerable Canon 100-400mm F4.5-5.6 IS lens.
I was fortunate enough to obtain a copy of this lens for review from Gary Farber of Hunt's Photo and Video, an online supplier of photography equipment. Gary and I were at the Outdoor Writers Association of America annual conference this past June, and I had emailed him in advance asking that, if he happened to have this lens in stock at the time of the conference, he bring it along in a Canon mount so that I might give it a try.
Taking it out of the box, I found that this new lens is built exactly like the EX series of Sigma lenses, which they tout at their professional line. It is solid, relatively heavy, and has the exact same exterior coating as previous EX lenses. The zoom function is very smooth, the lens hood easily attached and detached and the buttons for optical stabilization (OS), manual-auto focus, etc., are easily manipulated and well placed. Some have speculated that this lens doesn't have EX designation because Sigma is now saving that for lenses that maintain their maximum aperture throughout the focal length. I haven't a clue if that is true, but I can say that, after having owned a couple of Bigmas (the 50-500mm Sigma lens) that this new lens physically seems to be built to the same standards.
Gary let me take this camera into the field, and I was excited to get up early with the intent of doing some waterfowl and wildlife photography in the countryside surround the conference site of Bismarck, North Dakota. I ventured afield with friend and fellow wildlife photographer Tim Christie, and anticipated an excellent morning.
THE NUTS AND BOLTS
I brought an extra quick release mount with so that I could put the lens on my Bushhawk shoulder mount because I wanted to use this lens as I would any other lens in my arsenal, and that's how I primarily shoot -- hand held with the Bushhawk as an aid. When we pulled up to the first prairie pothole, I shut off the engine and Tim and I began photographing a small group of redhead and canvasback ducks. And something seemed terribly wrong almost immediately.
The Sigma's autofocus, although as quick to initially lock on as I'm used to from my Canon "L" series lenses, tended to easily lose focus. In a series of about ten shots of this group of slowly swimming ducks I found that half were completely blurred, a few were acceptable, and only one was nice and sharp. I also noted that when in AI servo focus mode, my 40D was slowed from its admirable 6.5 frames per second to about 2 or three fps. I was shocked. Maybe I was doing something wrong. I methodically checked all the camera settings, cleaned the lens and camera contacts, and moved on to the next photo subjects.
The next prairie pothole revealed a group of white pelicans. As they rose into the air I attempted what is routine for me, a burst of flight shots. Again the lens limited my camera to about 2 fps, and autofocus was less than reliable. A few ruddy ducks were swimming nearby, so I tested the lens on them. I could see the frame in the viewfinder going in and out of focus. A road sign nearby yielded the same results. My goal wasn't to shoot some static test shots, but to use this lens in the field as I would any lens, and judge it at the end of the day by evaluating my yield of "keepers." Yet, as interested as I was in giving this lens a thorough test, I was also interested in getting some good images for sale, so I swapped this lens for my trusty old Canon 100-400 and finished out an enjoyable morning with Tim.
Later in the day, and back at the motel, I tinkered some more with this lens. I found that on my 40D, in One Shot or AI Focus mode, my full 6.5 fps returned, and autofocus seemed more reliable. I slapped it on on a Canon 1D Mark III, and a 30D, and found that the maximum fps second was maintained in all modes when using either of these bodies. Now I was beginning to wonder whether or not it was the 40D body that was malfunctioning, instead of the Sigma lens.
I hunted down a couple other photographers at the conference, and tried the lens on their 40D bodies and discovered that in AI Servo mode, the new Sigma 150-500 limited every 40D body to about 2fps. This clearly is an incompatibility issue with this particular model of body, and although I expect Sigma will rectify it soon, I would urge anyone shooting a 40D to try the lens on the body before making a purchase. It is possible that the single lens I had was somehow defective, and I will be watching photography forums to see if anyone else has similar issues with this combination.
I would suspect that the less-than-reliable focus is also a part of this compatibility problem, because I later shot some images of vehicles speeding by, using a 30D, and found the lens to perform well. It maintained focus better in going away shots than in incoming shots, but I've found that to be the case with many lenses, including "L" series lenses from Canon. The speed of the autofocus is on par with that of the 100-400 Canon.
The optical stabilization function works well in both modes, although it is very noisy when compared to Canon's system. In fact, I'd have to say that the Sigma OS works better than the Canon IS (aside from the noise), at least when compared to the Canon 100-400, which has an older generation system.
So....up to this point we have a mixed review. The Sigma 150-500 has a good solid feel, seems well built, is laid out nicely, zooms smoothly, and has a good image stabilization function. Autofocus is fast and, on a Mark 1D and 30D, was accurate. If you're shooting a 40D, be careful, as their seems to be a compatibility problem that slows burst rate, and makes autofocus unreliable when shooting in the AI servo mode.
THE HEART OF THE ISSUE -- IMAGE QUALITY
Let's say Sigma fixes the compatibility issue and this lens works flawlessly on all camera bodies. (UPDATE -- Sigma has fixed what was a chipset problem in this lens, and new models are now compatible with the 40D). What about the heart of any lens? What about image quality?
Again, back at the hotel I mounted this lens on a tripod, and along with the Canon 100-400, I shot some boring images of flowers outside the building. I was excited to download them because I was truly hoping for Sigma to succeed with this lens where they had failed with the 50-500, which has never equaled the Canon in image quality.
I won't beat around the bush. The image quality of the Sigma 150-500 is better than what I recall from the 50-500, but still falls short of the Canon. Colors appear overly saturated at the same in-camera settings, and detail is lacking. Although the lens produces nice results that any amateur photographer would be pleased with, it does not produce professional results at 500mm, plain and simple.
As a test, I made sure the focus point (center only) was on one flower blossom. I shot the scene with both lenses, picked the sharpest of multiple images from each, and compared them. When using the Sigma, I shot it at both 400mm and 500mm, and shot both lenses at their fastest aperture. Why didn't I shoot the Sigma at other apertures? Because my intended use for this lens, like it is for the Canon 100-400, is for action wildlife photography. While I do care if the lens gets "sweeter" as you change apertures, if it doesn't perform wide open, then I'd never purchase it.
Let me repeat that the images produced by the Sigma are very, very nice. When comparing the best 400mm images from each lens, the Sigma images were nearly as good as Canon images. Having said that, remember I shot both at their maximum aperture. Had I shot the Canon at f/6.3 (the same as the Sigma's maximum aperture) I believe it would have pulled ahead, as this lens also "sweetens" above f/5.6. However, at 500mm and at f/6.3, the Sigma grew unacceptably soft.
To satisfy my curiosity, I even took a Canon image (taken at 400mm f/5.6) and interpolated it up in size so that the flower blossom would appear the same size as it does in the Sigma image (taken at 500mm f/6.3). Despite a process (interpolation) that can degrade image quality the "pseudo-500" from the Canon STILL was as good or better than the image from the Sigma. By the way, I did not change the saturation (you'll see a difference) and no post processing at all (including sharpening) was done. The in-camera sharpening was set at two. And no, one wasn't taken in the shade and the other in the sun, despite the difference in brightness. Compare the shadows in the image (the white blossom in left center is a good example) and you'll see that the angle of the sun was the same. I can't quite explain why the two lenses produced such different results, but both were acceptable in brightness. Note, though, that the Canon better retained detail in the whites, which is a very desirable result.
Here's the full frame, (reduced in size) with EXIF info, taken with the Sigma lens at 500mm.
Here's the full frame (reduced in size) with EXIF info, taken with the Canon 100-400 at 400mm, then interpolated UP in size so that the objects in the frame are the same relative size (that is, as if it were taken with a 500mm lens).
Plain and simple, at 500mm the Sigma just doesn't give the image quality I had hoped for, or that a Canon "L" series lens routinely delivers. Considering that you can even interpolate a Canon 400mm image up to an equivalent size as one taken with a the Sigma at 500mm and have as good or better image quality than that produced by the Sigma, even the Sigma's selling point of giving you an additional 100mm focal length quickly falls aside. The only thing that the Sigma did better, and only marginally, was in image stabilization, but that was counteracted by an annoying metallic whine.
If it sounds like I'm coming down hard on Sigma, I guess I am. Like many out there, I was hoping this lens would be on par with the Canon 100-400. I've been looking hard for a hand-holdable 500mm lens (are you listening Canon????), and I don't need an F4 aperture. Sigma had the chance to make a lens that would give Canon a run for their money. Instead, they delivered a lens that, while in many respects is quite nice, fails where it really counts -- image quality at 500mm. Sure, it is cheaper than the Canon, so what did I expect? Well, yes, it is cheaper, but not by much. If, by improving image quality, Sigma needed to sell it for about the same price as the Canon 100-400, they should have done so. They'd still have sold a ton, and they would have given the consumer a product that competes with Canon.
But until they adopt that philosophy, I have one word of advice for those of you who are considering purchasing this lens in a Canon mount -- pay the extra bucks and get the Canon 100-400. It is a much better lens.
I've received several emails about the above review, picking nits with what I wrote. Some correspondents have even sent stunning photos taken with the Sigma 150-500 at 500mm that I would be proud to have taken. BUT....and this is a big BUT....all of the really sweet images were taken at apertures of f/8 or f/11.
It doesn't surprise me that the lens
produced better results at a smaller aperture. Most lenses do. If you go
back and re-read my review, you'll note I was looking for a 500mm lens
that would perform at its maximum aperture. At f/6.3 this lens is already
marginal in lens speed -- having to use it at f/8 to get good
results just wasn't worth the investment FOR ME. Your shooting needs may
So...my conclusion remains unchanged. This is a very nice lens, suitable for amateur work, and it does what it is advertised to do. But it is NOT up to par with the Canon lens, or suitable for professional quality images. If you would like to try the Canon 100-400 before buying, consider a lens rental.
|COPYRIGHT NOTICE: All photographs appearing on this site are the property of Michael Furtman. These photographs are protected by U.S. copyright laws and have been registered with the U.S. Copyright office. They are not to be downloaded or reproduced in any way without written permission. By entering this site you accept these terms.|