Field Tests: 500mm Wars - Sigma vs. Nikon
Introduction
Post Date: March 22, 2017
This field test comparing the Sigma 500mm f4 DG OS HSM Sport to the AF-S Nikkor 500mm f/4E FL ED VR super-telephoto primes originally appeared as a series of blog posts in late 2016 and early 2017. This version contains no new information but should be in a more convenient format for most viewers.
In this extensive field test I compare the Sigma 500mm f4 DG OS HSM Sport prime lens to the latest Nikkor 500mm lens - the AF-S Nikkor 500mm f/4E FL ED VR. The bulk of the field-testing was performed in a "head-to-head" fashion comparing the two 500's, but I also compare some aspects of both 500s' performance to the Nikkor AF-S 400mm f/2.8E FL ED VR (with and without a 1.4x teleconverter) and to the Sigma 150-600mm f5-6.3 Sport zoom shot at 500mm. In short, my goal is to answer the following questions:
I. Which 500mm lens best meets MY needs as a wildlife photographer?
Note that the two 500's have extremely similar specifications but differ by about 40% in price (with the Sigma coming in at - quite literally - thousands of dollars LESS than the Nikkor 500).
Historically many professional and serious amateur wildlife photographers (including myself) have been biased against 3rd party lenses, including the offerings from Sigma. Even while the most open-minded of the 3rd party lens "skeptics" would grant that the Sigma lenses might be "good for the price", they really didn't stack up in an absolute sense (judged by image quality, AF performance, etc.) with the best Nikkors. My own eyes were opened (and my bias against Sigma's lenses started disappearing) when I was sent a copy of the Sigma 120-300mm f2.8 DG OS HSM Sport lens for testing way back in 2013. Long story short, I was blown away by the overall quality of that lens (the longer story about what I thought of the Sigma 120-300 can be found on my 13 August 2013 blog entry - right here).
If Nikon had a pro-quality DX camera at the time I would have kept (i.e., BOUGHT) the Sigma 120-300mm f2.8 Sport lens.
And then, of course, came the Sigma 150-600mm f5-6.3 Sport lens. I have commented extensively about how favorably I think of the Sigma Sport 150-600 and - long story short again - I tested it against virtually all competing zooms (the two appropriate Nikkors - the AF-S 80-400mm VR, the AF-S Nikkor 200-500mm lens, the original Tamron 150-600, and Sigma's own 150-600mm Contemporary) and kept only ONE of them for my own use - the Sigma Sport 150-600mm.
In short, Sigma is now trying VERY hard to "play with the big boys" and doing a really good job at it. At the beginning of this test I had no preconceived notion over which of the two 500's would best meet my needs - I DID think it was possible that the Sigma 500 would equal or even outperform the Nikkor 500. During this testing I compared both 500's lenses shot "alone" (without a teleconverter) as well as with their respective teleconverters.
II. How do BOTH 500's compare to Nikon's excellent 400mm f2.8E VR (both shot native and with the 1.4x (TC-14EIII) teleconverter?
III. How do BOTH 500's compare to the Sigma Sport 150-600mm zoom shot at 500mm? Or, said another way, "Does it really make sense anymore to fork out the big bucks for the big and fast primes?" ;-)
The ultimate goal of all this field-testing for me was to end up with the best quality "kit" for my own wildlife photography needs. I had NO plans to keep BOTH 500's, but it was possible at the onset of the test that I would keep NEITHER of them (if I find there is little real net advantage of the "best of the 500's" compared to my 400mm f2.8E). Along the way in this test I discovered a LOT that I think will help others decide which of these lenses will be best for them.
As with all my field-tests, this review contains information from systematic and quite rigidly controlled "tests" through to my observations after extensive periods of "just shooting" the two 500's.
Caveats, Qualifiers, and Limitations of My Results:
Everything I do and write about during my field-testing of the two 500mm lenses comes from the perspective of a WILDLIFE photographer. There will likely be LOTS of what I have to say that will apply to photographers of other genres (e.g., sports photographers), but I readily admit my wildlife photographer "bias". And, I tested ONE copy of each lens only. While one would like to assume that lenses costing over $5000.00 (in any currency) that they are built with a high degree of quality control and there is little between-sample variation in quality, it's possible that the results I obtain MAY differ a little from what others find with their own copies of the lenses.
I test my gear quite extensively in an effort to discover how it will perform for me (using my own shooting style) in a field situation. I don't do these tests for profit, but simply to understand how the product in question will work for me in the field and thus so I can understand how I can use the product to better create images that I can sell. I test gear under field conditions only (no lab work) and use the same techniques I'm likely to use when I'm shooting the particular item in the field. While I do some of my testing very methodically, much of it is pure "field shooting". Aside from when I'm doing some lens focus-tuning, I do not shoot images of targets under rigidly controlled lab conditions - I shoot images of wildlife (or "proxies", such as my Portuguese Water Dogs) in the field. It's not critical to me to produce results that are generalizable or that are rigorous enough to be published in a peer-reviewed journal - I care about how I can use the gear in the field and how to get the results I need to sell images! While some "lab tests" have a real-world correlate that translates into a limitation in the field, I find an increasing number of tests quite esoteric and the "differences" between two products is real only in a statistical sense (and has little or no real-world correlate in producing a quality image, which is NOT a pure science). So you won't see any MTF charts in this comparative field test!
So, in short, these are "real world" field tests from the perspective (and biases) of a professional wildlife photographer. They help me understand my gear. And, based on the web traffic each of my reviews receives and the email I receive from those who read them, they appear to help others. That's gratifying.
Statement of Objectivity:
Hey, I'm a skeptic of what I read online myself. And I always wonder about the "corporate influence" on so-called product reviews (which are often little more than spec spews). For the record, I have received NO differential "perqs" or incentives (real or implied) to push my results one way or the other - I performed this test without any form of "sponsorship" by Sigma or Nikon. In this particular head-to-head field-test I asked neither Nikon or Sigma to supply me with a lens free of charge for testing purposes. In other words, I BOUGHT the two lenses and will sell one or both of them at the end of the test.
Why the Big Effort on These 500mm Lenses?
This is a great question - this field test ate a TON of my time over the winter of 2016/2017...so why did I do it? To many wildlife shooters (of any brand) the 500mm lens is their absolute "dream lens". With both Nikon and Canon their top 500's are the smallest, lightest, and cheapest of their "big 3" super-telephotos (that most think of as the 400mm f2.8, the 500mm f4, and the 600mm f4). I have to admit that for my own uses I have always preferred the 400mm f2.8 over the 500mm f4, but as airlines tighten up their weight and travel restrictions (particularly for carry-on bags), the appeal of the smaller, lighter 500mm lenses (compared to the 400mm f2.8) increases. Over the past year or two the two "Japanese giants" of professional DSLR photography (yep, Nikon and Canon) have increased their lens prices to the point where buying a 500mm lens is a huge financial commitment (or, for some, has become totally out of the question!). So...now that Sigma has charged in with a 500mm lens that is about 40% cheaper...well...there are going to be a LOT of wildlife shooters wondering if the lens can match their much more expensive competitors. And you can put me in that group - as one who already owns the wonderful 400mm f2.8E VR I have a hard time justifying buying a Nikkor 500mm f4E VR, but if the Sigma is virtually as good overall...well...it just may end up in my kit. So this extensive field test will help ME make an expensive and important decision - and along the way may help others make similar decisions.
Some Terminology and Abbreviations:
To save typing and reading time, here's some terms and abbreviations I'll regularly be using in this review:
• Sigma Sport 500mm f4 (or just "Sigma Sport 500" or even "Sigma 500") = Sigma 500mm f4 DG OS HSM Sports prime lens
• Nikkor 500mm f4E VR = AF-S Nikkor 500mm f/4E FL ED VR prime lens
• Nikkor 400mm f2.8E VR = AF-S NIKKOR 400mm f/2.8E FL ED VR prime lens
• Sigma Sport 150-600 = Sigma 150-600mm F5-6.3 DG OS HSM | S zoom lens
• Native (as in "shot native") = without a teleconverter
• TC = teleconverter
A Few Words on Funding and Donations
Much to the chagrin of my accountant, I don't accept donations (or "tips") for my work. However, if you would LIKE to help motivate me to continue this work I am very open to (and would highly appreciate) that in lieu of giving ME a donation you send it to the science-based conservation conservation group I work most closely with. This group - The Raincoast Conservation Foundation - does excellent work (including funding of peer-reviewed scientific studies) and has been a key player in saving much of British Columbia's Great Bear Rainforest.
Donations can be made in two ways:
• Directly to Raincoast at: www.raincoast.org/donate/. If you choose this route YOU will receive the tax receipt.
• Send ME the donation via PayPal at payments@naturalart.ca. If you choose this route I will receive the tax receipt. Note that the donor will receive full verification of the donation to Raincoast.
A Quick Directory To What's Coming:
• Physical Characteristics
• AF Tuning Values...
• Optical Performance
• 1.4x Teleconverter Performance
• Stabilization and "Hand-holdability"
• Autofocus Performance
• Musings from Just Shootin'
• Wrap-up and My Lens Choice
Physical Characteristics
Here are my impressions of the physical characteristics of the two lenses. It is not intended as a "spec spew" although I will point out a few specs that seem to have been overlooked by many. I've previously stated that the specs on these two lenses are very similar - after having them in my hands for several days this statement seems almost like an understatement - these two lenses are incredibly similar overall (at least in a physical and "spec sense"). Those wishing to review the detailed specs should go HERE for the Nikon 500mm f4E.
To get a feel for the specs of the Sigma Sport 500mm you have to visit both the spec list on dpreview.com (HERE) and the spec listing on the Sigma Photo website (HERE).
I. A FEW Specification Highlights
Here's a few of the more easily missed or overlooked specs that may be important to some photographers...
• Electromagnetic Diaphragm? Yes for both lenses. But note that the Canon version of the Sigma 500mm f4 does NOT have an electronic diaphragm. Why is having an electromagnetic diaphragm significant? Electromagnetic diaphragms ensure stable exposures throughout a high-speed burst of shots (normally only needed when you're shooting at about 8 fps or higher). This is a good thing. Both lenses have it.
• Minimum Focusing Distance? 3.6 meters (11.8 feet) for the Nikon 500mm f4E and 3.5 meters (11.5 feet) for the Sigma 500mm f4.
• Optical Stabilization Systems? 2-mode VR system for the Nikon 500mm f4E (Normal vs. Sport settings) and 2-mode OS system for the Sigma 500mm f4 (OS1 = hand-held and OS2 = Panning). The stabilization system of both lenses are purported to give a 4-stop advantage in vibration reduction (over non-stabilized lenses). Note that I will be saying much more about how these compare (both theoretically and in practice) in my section below on "Stabilization and Hand-holdability".
• Fluorite Lens Elements? Yes for the Nikon 500mm, but NOT the Sigma 500mm. Why is this important? Primarily for weight saving (and it just so happens that the largest and heaviest lens elements on the Nikon are the fluorite ones).
• Number of Diaphragm Blades? 9 for both lenses. Why is this important? Having more diaphragm blades impacts on the quality of the out-of-focus (or OOF) zones. This adds an expense compared to having 8 or fewer diaphragm blades.
• Country of Manufacture? Japan for BOTH lenses.
• Environmental Sealing? Yes for both.
II. What's in the Box?
OK...this might seem like a bit of silly thing to talk about (though not as silly as those ludicrous "unpacking videos" you'll find online!) but it does reflect a "big picture" difference I'm noticing between Nikon and Sigma: one of them seems focused on function and value and the other is...uhhh...a bit more driven by history and tradition (and not too receptive to change). What do I mean? Here's an example: The two lenses come with pretty much the same bits in the box, including carrying case, lens covers, et cetera. And Nikon has spent a bundle on giving you an absolutely beautiful carrying case that looks like high-end luggage. It includes internal "sculpting" to perfectly match the lens. Almost a work of art! Sigma, in contrast, includes a padded cordura case that holds the lens securely but has room for other accessories inside, like a pro body and other bits. And, the Sigma case has backpack style straps on it. In short, the Sigma case is actually quite functional and I can see a lot of owners using it as airline carry-on or even in the field. I'll definitely use that case. I can't speak for everyone else, but I can't imagine using the Nikon case for anything but putting the lens in when I sell it and need to ship it to someone else (which is the ONLY thing I ever did with the similar cases that came with my old and long-gone Nikkor 400mm f2.8G VR and my Nikkor 600mm f4G VR). Nikon case: very nice and classy but pretty useless. Sigma case: not too pretty, but kinda useful and functional.
III. Some Physical/Design Differences BETWEEN the Lenses
While there are an amazing number of similarities between the lenses, there are a few physical differences between them. Here's what I've noticed so far:
• AF Activation Buttons ((AKA "AF Function" Buttons on the Sigma 500mm): These are the four round buttons found around the lens near the distal (or far) end of it. With BOTH lenses you can "program" the buttons to do 3 things: AF-On (activate the autofocus), AF-L (lock the focus when in AF-C or continuous focus mode), or return the lens to a preset focus distance (Memory Recall). BUT, the two lenses have the buttons positioned differently - on the Nikon 500 the four buttons are along the top, bottom and side of the lens at 90º intervals. Same with the Sigma, but they are offset from being exactly on top, bottom, and on the sides, with the net result that they fall directly UNDER my thumb (where you want them!) when I'm hand-holding the lens horizontally OR vertically. Smart. Functional. Oh, and BTW...with both lenses the AF-L function of these buttons can be used to change AF area mode (assuming you're using a Nikon body that offers this functionality).
• Focus Limiters: Both lenses have a switch to limit focus to specific distance ranges. This can help speed up initial acquisition of focus by preventing the lens from winding all the way in (to closest focus) when the subject isn't too close to you. With the Nikon 500mm there are TWO positions on this switch - "Full" and "Infinity to 8m". On the Sigma there are THREE positions on this switch - "Full", "10m to Infinity", and "3.5 to 10m". I'm thinking I'll quite like the Sigma 3 focus delimiter system (and the distances CAN be customized using the USB dock and Sigma Optimization Pro software - see immediately below).
• Customization Switch: This switch is found on the Sigma 500mm only (there's nothing comparable on the Nikon 500) and is used to switch between default lens settings or one of two custom settings you have set up for the lens using the optional USB Dock and free Sigma Optimization Pro software (which can also be used to update the lens' firmware, including changes to the autofocus algorithm). There are many parameters that you can change that affect either AF or OS performance and then assign to one of the two custom functions (as indicated on the switch). I'll leave a detailed discussion about this whole issue of user-performed lens customization (and firmware updates) to a detailed blog entry in 2017, and for now all I'll say is that I think this is a fantastic feature (and it's unique to the Sigma).
• Functioning of Lens Collar: As you'd expect with any "big" lens both lenses come with rotating tripod collars and "stock" tripod feet. Of course (and don't ask me why), both lenses have stock tripod feet that are NOT Arca-Swiss compatible and you have to add either a lens plate OR a full replacement foot to use the tripod foot with an Arca Swiss tripod head. So both are equally bad in this respect! But let's go back to the rotating lens collar because THEY differ a little in function. The Nikon 500mm lens collar rotates smoothly and continuously with no detents. In contrast, the Sigma 500 collar has two modes - with detents (at 90º degree increments) on or continuous (sans detents). Smart. And, if you compare how smoothly the two lenses rotate when the Sigma has the detents turned off...well...I have to say the Sigma rotates much more smoothly (as though it's on bearings and with no "play" whatsoever).
• Design/Appearance Differences: Both lenses are in a matte black finish and Nikon has chosen to gradually increase the width of the lens from the proximal portion that connects to the camera to the distal portion. The Sigma differs in that the lens stays fairly constant in diameter until past the focusing ring, at which point it jumps quickly in diameter. To be honest I can see no functional difference in the lenses associated with this differing taper. Overall I'd describe the Nikon 500mm as "elegant" looking and the Sigma looks more "business-like" (or more clearly based on function and not form). Kind of modern design with the Nikon vs. Soviet Bloc era design with the Sigma! ;-) Purely eye-of-the-beholder (personal preference) stuff...
IV. Perceived Build Quality
Ok...here we go! This is one of those nebulous characteristics that defies definition but we all have a feeling for (reminds me of Pirsig's - from Zen and the Art of Motorcycle Maintenance - almost non-ending search for what "quality" means!).
But this is going to be a short section: both lenses seem absolutely EXCELLENT in build quality and I can't see anything on them that separates them from one another (other than perhaps how smoothly the Sigma lens collar rotates compared to the Nikon). There are many other competing lenses where build quality differences are instantly apparent such as when one compares the Sigma Sport 150-600mm zoom to the Nikkor AF-S 200-500mm VR or the Tamron 150-600mm zoom - in those comparisons the Sigma Sport was definitely a head and shoulders above both the other lenses in build quality. But the two 500's I'm comparing here...both seem built for long-term professional use. In my mind the build-quality of these two Japanese-made lenses is a complete saw-off. Of course, this is something that tends to show most only after several years of use...so we'll have to wait a while to get THAT report! ;-)
V. Tale of the Tape - Size Differences
OK...not MUCH difference here. If you measure the Sigma lens from where it meets the camera body to the end NOT including the hood you'll find it's 380mm (14.96") long. The Nikon 500mm is 388mm (15.25") long. The lightweight Sigma hood is slightly shorter than the Nikon hood, so when those are added on the Sigma lens measures 522 mm (20.55") to the Nikon's 545 mm (21.45"). So the Sigma is almost an inch shorter when in "shooting" mode.
VI. Tale of the Scale - Weight Differences
I weighed the two lenses several different ways. First, I stripped everything off them, including the next-to-useless stock tripod feet. Stripped down the Nikon 500mm came in 324 gm (.714 lb) LIGHTER than the Sigma Sport 500. The Nikon 500mm f4E came in at 2906 gm (6.4 lb) and the Sigma Sport came in at 3230 gm (7.12 lb).
What happens when you put the "required" things on them and them weigh them as they are when you're actually shooting with them (i.e., how do the shooting weights compare?). Well...the hoods are almost equal in weight (286 gm for the Nikon hood and 290 gm for the Sigma hood). I replaced the stock tripod feet (foots?) on both of the lenses with 3rd party feet - in the case of the Nikon 500mm f4E I used a Really Right Stuff LCF-14 (it fit fine and was long enough to balance camera bodies of very different weight - D5 vs. D500 without battery grip - just fine). On the Sigma I used the long Jobu replacement foot that was built for the Sigma Sport 150-600 (again, it fit fine, but in this case it's probably longer than needed to accommodate the balancing of any Nikon body, regardless of weight - info for the Jobu foot HERE). NOW the Nikon 500mm f4E weighed 3320 gm (7.32 lb) and the Sigma Sport 500mm weighed 3644 gm (about 8.03 lb). The difference? 324 gm (.714 lb) again! So...no matter how you look at it...the Sigma Sport 500mm weighs almost 3/4 of a pound MORE than the Nikon 500mm f4E.
Looking for a reference? Well...consider the 500mm lens that the "new" Nikon 500mm f4E replaced (i..e, the 500mm f4G). I weighed one I was recently was shooting with and its shooting weight (with hood and that same RRS LCF-14 replacement foot) came in at 4166 gm (or 9.18 lb). So...the NEW Nikkor 500mm f4E is 1.86 lb LIGHTER than the "old" 500 it's replacing, and the Sigma Sport is about 1.15 lb LIGHTER than the old Nikon 500. So in total weight I'd describe the Sigma Sport 500mm as very good, and the Nikon 500mm f4E as even better!
Will the weight difference between the Sigma Sport 500mm and the Nikon 500mm f4E make a difference in the field? For some - definitely. It could make the difference in whether or not some users can hand-hold the lens effectively (or how LONG they can effectively hand-hold the lens). For others...it won't be that significant...and the Sigma IS still a lot lighter than the OLD Nikon 500mm f4G. Note that at the end of November I shot with the "old" (and heavier) Nikon 500mm f4G for a full week and found even this "older and heavier" lens quite easy to hand-hold. Of course, ultimately lens/camera balance and the effectiveness of the image stabilization system will play a huge role in determining how slow a shutter speed any user can hand-hold these lenses at. Of course, I will be testing that in the coming days.
One final comment on lens balance: I find BOTH of these lenses to be very well balanced. I placed each of them on a gimbal tripod head and the distance between the balance point and the rear of my D5 to be virtually identical (to the millimeter). That "tilt test" confirmed my feeling that the two lenses felt VERY similarly balanced when I was hand-holding them when attached to my D5.
So...where does this leave us right now? Well...a definite and noticeable edge to the Nikon 500 over the Sigma 500 in lens weight. A dead heat in build quality. But an edge to the Sigma in lens features.
AF Tuning Values...
While I am not one who is fixated on endlessly fine-tuning the autofocus system on my cameras and lenses (I tend more towards a "...if it ain't broke don't fix it" attitude towards AF tuning), I thought it would be prudent to check the focus tuning on the lenses I am evaluating and comparing in this "500mm Wars" series (and to do so BEFORE serious lens testing). For consistency's sake, I used the EXACT same protocol in AF Tuning all the lenses in this test which, in this case, means I used the Nikon D5's and D500's "automated" AF tuning features and followed the 8 "fluid guidelines" I described in my 27 April 2016 blog entry entitled "The Nikon D5/D500 and Automated Lens AF Tuning". Please note that I will have further comments on Sigma's proprietary multi-distance (and multi-focal length) AF tuning in a blog entry in 2017.
Here's a quick list of the camera/lenses and camera/lens/TC combinations I "tuned" for the field tests:
• Nikon D5 with Sigma 500mm f4 Sport (native AND with TC-1401 teleconverter)
• Nikon D5 with Nikkor 500mm f4E VR (native and with TC-14EIII teleconverter)
• Nikon D5 with Sigma Sport 150-600mm @ 500mm (native)
• Nikon D5 with Nikkor 400mm f2.8E VR (native and with TC-14EIII teleconverter)
• Nikon D500 with Sigma 500mm f4 Sport (native AND with TC-1401 teleconverter)
• Nikon D500 with Nikkor 500mm f4E VR (native and with TC-14EIII teleconverter)
• Nikon D500 with Sigma Sport 150-600mm @ 500mm (native)
• Nikon D500 with Nikkor 400mm f2.8E VR (native and with TC-14EIII teleconverter)
And here's a short list of the values for a few of the variables in the AF Tuning:
• Distance to Target: For all combinations with the D5: 25 meters. For D500: 35 meters.
• Sample Size: 24 measurements for EACH camera/lens combination. Up to 4 obvious outliers removed from each sample, with most averaged values based on a sample size of 21 to 23 measurements. In almost all cases the measured values were quite stable and outliers were easily identifiable (and relatively rare).
After inputting the tuning values obtained (immediately below) I took test shots of objects with continuous foregrounds and backgrounds and then visually examined the result - all to ensure that the measured tuning values reflected reality (and didn't make the focus worse!). And they did reflect reality! ;-)
Resultant AF Tuning Values - here's the list:
• Nikon D5 with Sigma 500mm f4 Sport (native) = +3
• Nikon D5 with Sigma 500mm f4 Sport plus TC-1401 teleconverter = +1
• Nikon D5 with Nikkor 500mm f4E VR (native) = -1
• Nikon D5 with Nikkor 500mm f4E VR plus TC-14EIII Teleconverter = -8
• Nikon D5 with Sigma Sport 150-600mm @ 500mm (native) = -3
• Nikon D5 with Nikkor 400mm f2.8E VR (native) = 0 (zero)
• Nikon D5 with Nikkor 400mm f2.8E VR (native) plus TC-14EIII Teleconverter = +5
• Nikon D500 with Sigma 500mm f4 Sport (native) = +4
• Nikon D500 with Sigma 500mm f4 Sport plus TC-1401 teleconverter = +2
• Nikon D500 with Nikkor 500mm f4E VR (native) = -1
• Nikon D500 with Nikkor 500mm f4E VR plus TC-14EIII teleconverter = -4
• Nikon D500 with Sigma Sport 150-600mm @ 500mm (native) = -7
• Nikon D500 with Nikkor 400mm f2.8E VR (native) = 0 (zero)
• Nikon D500 with Nikkor 400mm f2.8E VR (native) plus TC-14EIII teleconverter = +7
Please note that these values are absolutely unique to MY cameras and lenses...they in no way represent recommended values for your own gear!
Optical Performance
I. INTRODUCTION
One of the most critical variables influencing any lens purchasing decision is the optical performance of the lens. Optical performance becomes particularly critical when deciding on which super-telephoto prime lens to buy - not only are thousands of dollars at stake, but the whole rationale for considering a super-telephoto lens (over a zoom lens covering the same focal range) is the belief that image quality will be of the highest quality.
In this section I describe the results of head-to-head field-testing of the optical quality of the "new" Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR prime lenses. I compared the lenses over a wide range of apertures when shot native (sans teleconverters), when shot with their respective 1.4x teleconverters, as well as against two other lenses - the Nikkor 400mm f2.8E VR (with and without the TC-14EIII (1.4x) teleconverter and the Sigma Sport 150-600mm zoom @ 500mm. I tested the lenses at 3 different distances - with close subjects (7 meters or about 23'), with mid-distance subjects (30 meters or about 100 feet), and distant scenes (1.95 km or about 1.2 miles to subject). I chose these distances as they represent the sort of distances I work with when shooting small mammals like squirrels or mid-sized birds (the 7 meter distance), larger mammals like bears (the 30 meter distance), and what I often refer to as "animal in landscape" (or animalscape) distances. I may add one more distance (100 meters to subject) to the mix, but - based on the results you'll soon hear about - I don't think there's much point!
Note that I approach sussing out lens optical/image quality a couple of ways. The first is examining the "theoretical maximum quality you could EVER get in the field". For this comparison I capture images using a high degree of control (much higher than I would use for about 99% of my wildlife photography). In this case that "higher degree of control" meant that I captured the images discussed today using:
• A firm tripod (Jobu Algonguin - info HERE),
• a firmly tightened down gimbal tripod head (Jobu Heavy Duty MkIV - info HERE),
• Live View (contrast detect AF), mirror-up, and electronic front shutter curtain,
• a MC-20 Cable release (to focus and to trigger the shutter),
• and with the VR or OS system OFF for all lenses.
A few other image capture notes: For the short camera-to-subject distances (7 & 10.5 meters) I used both a Nikon D5 and D500. For the 30 meter distance I used only a Nikon D5. For the 1.95 km distance I used both a D5 and D500 AND I also added in images shot with a D800e (primarily to examine edge sharpness on distant scenes). I captured images from wide open (f4 for the 500m lenses) up to f16. I used 1/3 stop increments from wide open to f8, then single stop increments to f11 and f16. At each aperture I captured two images separated by an interval of about 10 seconds (to allow any vibration associated with shutter movement to dissipate) and re-focused between successive frames (and this step WAS necessary...there were instances where one of the two shots for a particular lens/aperture/distance combination slightly missed focus and consequently was slightly sharper).
The second way I look at lens image/optical quality (and overall usefulness of a lens) is how find out how much of the "theoretical" image quality can actually be realized in a field setting (where one is commonly hand-holding lenses, or shooting moving subjects, etc.). This realized image quality is influenced by other lens characteristics, including lens weight, balance, effectiveness of the stabilization (OS or VR) system and, of course, the effectiveness of the autofocus system. To get a better handle on this realized image quality in the coming days and weeks I will be testing the VR/OS effectiveness, the AF system, and spending time "just shooting" the lenses as I would during normal would when working with wildlife.
Note that today's comments on image quality are based primarily on the "theoretical maximum image quality" that is possible to extract with each of the lenses. Comments on what I have been actually "realizing" when "just shooting" will come later.
Finally, please be aware I am comparing only ONE copy of each lens and there can be some variation between copies of the same model of lens (I have always believed that between-sample quality variation is lower on high-end prime lenses than lower-priced consumer or "enthusiast" lenses...but that does NOT mean that there can not be some variation between samples in super-telephoto lenses).
While my primary focus in this testing was to compare the two 500mm lenses, ultimately I was trying to answer 5 questions:
1. How does the image quality of the Sigma Sport 500mm f4 compare to that of the Nikkor 500mm f4E VR over a wide range of aperture settings and at several distances?
2. How does the image quality of the Sigma Sport 500mm f4 compare to that of the Nikkor 500mm f4E VR over a wide range of aperture settings and several distances when each is paired with their 1.4x teleconverter (the TC-1401 and TC-14EIII, respectively)?
3. How does the image quality of both of the 500mm f4 lenses compare to images captured from the same position using the Nikkor 400mm f2.8E VR lens and then UPSIZED (or UPSAMPLED) in Photoshop to match the subject dimensions of the images shot with the 500mm lenses?
4. How does the image quality of both of the 500mm f4 lenses compare to images captured from the same position using the Nikkor 400mm f2.8E VR lens PLUS the TC-14EIII (550mm) and then DOWNSIZED (or DOWNSAMPLED) in Photoshop to match the subject dimensions of the images shot with the 500mm lenses?
5. How does the image quality of both 500mm lenses (AND the upsized and downsized 400mm shots) compare to images captured from the same position using the Sigma Sport 150-600mm f5-6.3 zoom lens shot at 500mm?
II. The KEY TWO SENTENCE SUMMARY: Just The Sigma Sport 500mm f4 vs. The Nikkor 500mm f4E VR
I have NEVER tested any two competing lenses that are so absolutely similar in image quality (at all distances, apertures, and with or without teleconverters) than the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR. Image sharpness, quality of the out-of-focus zones, and the progression in increasing sharpness from wide open through to about f5 (where both lenses approach maximum sharpness) is virtually identical between my copies of these two lenses.
III. The FULL EXECUTIVE SUMMARY
I compared image quality of four lenses plus various lens/teleconverter combinations over a wide range of apertures and at 3 different distances to subject: 7 meters, 30 meters, and 1.95 km. Surprisingly, I could find absolutely NO consistent differences in image/optical quality (in either sharpness or the quality of the out-of-focus zones) between the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR. This was true at all apertures and camera-to-subject distances tested and with all cameras tested (D5, D500 and D800e). Both 500mm lenses showed slight image softness (across the entire frame) when shot wide open (@f4), but both increased in sharpness at f4.5, and by f5 both were approaching maximum sharpness. Both lenses showed very good edge sharpness (and again it was virtually identical between the lenses). The optical similarity of these two lenses when shot "native" (without a teleconverter) was absolutely stunning - if I had not carefully keyworded the images (signifying which image was shot with which lens) it would have been impossible for me to determine which image was shot with which lens.
When the most current 1.4x teleconverters (the Sigma TC-1401 and the Nikkor TC-14EIII) were added to the lenses the remarkable optical similarity continued - I could detect no consistent differences in quality (again both in sharpness and the quality of the out-of-focus zones) between the lenses paired up with their TC's. Both lens and TC combinations were quite soft when shot wide open (i.e., at f5.6) but sharpened up somewhat by f6.3 and more by f7.1. Both were maximally sharp by f8 (stopped down a full stop from wide open when teleconverter attached). Personally I would not shoot either of these lenses wide open with their teleconverters attached. Speaking subjectively (and after looking at thousands of images shot with the 500's and with the Nikkor 400mm f2.8E VR with their teleconverters) it is my opinion that both the Sigma and Nikon 500mm lenses experience MORE image degradation when paired with their respective 1.4x TC's than the Nikkor 400mm f2.8E VR does.
How do images shot with the Nikkor 400mm f2.8E VR and then upsized in Photoshop (using the Preserve Details algorithm) to match the Sigma and Nikkor 500mm images in magnification compare in image quality to the Sigma and Nikon 500mm images? Not well. The upsized 400mm images were softer (when viewed at 100% magnification) than the images captured with either 500mm lens (at any aperture, including f4 on the 500's). Additionally the upsized 400mm shots showed excessive contrast (and thus appeared "harsher" than the 500mm images).
How do images shot with the Nikkor 400mm f2.8E VR plus TC-14EIII (550mm) and then downsized in Photoshop (using the Bicubic algorithm) to match the Sigma and Nikkor 500mm images in magnification compare in image quality to the Sigma and Nikon 500mm images? Quite well. At all distances images shot with the 400mm plus TC-14EIII (and then downsampled) were very comparable in sharpness AND in the quality of the out-of-focus zones to images captured with both 500mm lenses at apertures of f4 through f5. However, by f5.6 (and thereafter) the images captured using both 500mm lenses were slightly but noticeably sharper than images than the downsampled 400mm plus TC images.
How do images shot with the Sigma Sport 150-600mm (@500mm) compare in quality to the Sigma and Nikkor 500mm images and to the "digitally altered" 400mm f2.8E VR images? Pretty well for a zoom, but it ain't no prime! In other words, at ALL overlapping apertures the 500mm lenses were invariably sharper and with very easily seen "smoother and more buttery" out-of-focus zones (this was true at all distances and all apertures) than the Sigma Sport 150-600mm (@500mm). The Sigma Sport 150-600mm (@500mm) images WERE noticeably sharper than images shot with the 400mm f2.8E VR images that were upsized in Photoshop. However, images shot with the 400mm f2.8E VR combined with the TC-14EIII (550mm) and then down-sized to the magnification of a 500mm lens were sharper than any of the Sigma Sport 150-600mm shots at all apertures (though admittedly at f8 the images were quite comparable).
IV. MORE DETAILS:
Please refer to Appendix 1 for many more details on both the procedures used and the results I obtained at each test distance. There are a number of small gems contained within the appendix, but the primary conclusions have already been mentioned in the EXECUTIVE SUMMARY. But the detail-oriented types may enjoy the info!
V. CONTEXT...AND A FEW FINAL COMMENTS ON OPTICAL PERFORMANCE
For me there is one almost remarkable result coming through (repeatedly) in these field tests on optical performance: That when shot under conditions where you can extract close to the maximum optical performance of each of these two lenses, the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR are almost "optical clones" of one another. I have never field-tested two competing lenses that were so similar optically.
From my perspective the next HUGE question is this: "How much of this optical performance can you expect to fully realize when shooting the lenses under less controlled conditions, i.e., when shooting them more like you would when shooting wildlife?" It's my view that both autofocus performance and optical stabilization performance and "hand-holdability" play a huge role in how much of the "theoretical" optical performance you can actually realize in a real field setting. To get at this realized performance I'm going to do three things: test the AF systems, test the optical stabilizations systems, and JUST SHOOT with both of them in the field. Of course, I've already started the "Just shooting" phase of the exercise...and so far both lenses are looking pretty darn good...check out this action shot captured with the Sigma Sport 500 f4:
• The Joy of Running (Sigma Sport 500mm): Download 2400 pixel image (JPEG: 0.8 MB)
1.4x Teleconverter Performance
This section clarifies and expands upon on comments I made in the section above on the performance of the two 500mm lenses with their respective teleconverters (or TC's). Please note that these comments are limited to the two 500mm lenses with their 1.4x teleconverters and do not apply to their 2.0x teleconverters - I have not and will not be testing those. It's my view that at this point in time (and with the autofocus systems currently available to us) 2x teleconverters have very limited usefulness in a field-setting on f4 lenses. I have found that 2x teleconverters can produce excellent results on f2 and f2.8 lenses (and they have full autofocus capabilities with those lenses) - so in the case of Nikon the 200mm f2 VR (any version), the 300mm f2.8 VR (any version), and the 400mm f2.8 VR (both E and G versions) can produce excellent results when paired with the TC-20EIII.
I. SOME CLARIFICATION OF COMMENTS ON TC PERFORMANCE
In the section above I said a few things about the optical quality of the two 500's when paired with their TC's, including:
"When the most current 1.4x teleconverters (the Sigma TC-1401 and the Nikkor TC-14EIII) were added to the lenses the remarkable optical similarity [between the two 500's] continued - I could detect no consistent differences in quality (again both in sharpness and the quality of the out-of-focus zones) between the lenses paired up with their TC's. Both lens and TC combinations were quite soft when shot wide open (i.e., at f5.6) but sharpened up somewhat by f6.3 and more by f7.1. Both were maximally sharp by f8 (stopped down a full stop from wide open when teleconverter attached). Personally I would not shoot either of these lenses wide open with their teleconverters attached. Speaking subjectively (and after looking at thousands of images shot with the 500's and with the Nikkor 400mm f2.8E VR with their teleconverters) it is my opinion that both the Sigma and Nikon 500mm lenses experience MORE image degradation when paired with their respective 1.4x TC's than the Nikkor 400mm f2.8E VR does."
Clarification: I stand by this comment, but want to make it clear that this comment was NOT intended to convey anything negative about the image quality of either of the two 500's with their 1.4x teleconverters. In my view that performance (of either 500 plus TC) doesn't match that of the 400mm f2.8E VR with the TC-14EIII, but the 400 plus TC should be considered the "reference standard" for teleconverter performance - I feel that it's the absolute BEST quality you could ever expect out of any teleconverter/lens combo - it works phenomenally well! You can get VERY good results out of the Nikkor 500mm f4E VR plus the TC-14EIII AND the Sigma 500mm f4 Sport plus the TC-1401. If anyone's experience with teleconverters is limited to using them with zoom lenses it is likely they would to be BLOWN AWAY by the image quality they can obtain with either of these two 500mm primes plus their TC.
Here's a few sample images to illustrate what I mean - both were shot with the Sigma 500 plus TC-1401:
• Red Squirrel - Nikon D5, Sigma 500, TC-1401: Download 2400 pixel image (JPEG; 1.17 MB)
• Red Squirrel - Nikon D500, Sigma 500, TC-1401: Download 2400 pixel image (JPEG; 1.3 MB)
II. SOME ADDITIONAL COMMENTS ON 1.4X TELECONVERTER PERFORMANCE
Here are some further thoughts on TC performance:
A. Comparative Image Quality? I'm STILL finding that the two 500mm lenses shot in combination with their respective 1.4x TC's are virtually identical optically!
B. Autofocus? Subjectively, both 500's slow down slightly in initial image acquisition when their 1.4x TC is used. The slowdown isn't dramatic, but rather than almost instantly "snapping" into focus when shot native (which both lenses do very well), they sort of smoothly "slide" into focus! I have no real way to measure the slowdown in focus acquisition, but I'd estimate it at about 30% slower (than if the TC isn't in place). Both lenses still focus FAST. While I haven't had the opportunity to test focus-tracking on fast moving subjects with either lens plus TC in a field setting yet, based on what I've been seeing with slowly moving subjects (e.g., walking dogs), I'd be surprised if either combination struggled much with most bird-in-flight shots (probably not good to use them for full-frame shots of swallows in flight, but I doubt you'd have ANY problem with gulls, ravens, hawks, eagles, et cetera).
Note that I HAVE compared the initial focus acquisition using all 55 selectable focus points on both the D500 and D5 and my comments immediately above about the slight slowdown in initial acquisition of focus applies equally to ALL focus points (i.e., I saw no difference in the slight slowdown between any of the focus points on the AF array). Please note that here I am referring ONLY to speed of focus acquisition here - NOT accuracy of focus (I will discuss that in my Autofocus segment of this 500mm wars series, which is coming within the next week).
C. Are the TC's Worth The Money? Well...this is the type of question that is hard to give a "universal" answer to, but I sure think they're worth the money. And, I will be keeping and using (likely on a fairly frequent basis!) the TC with the 500 lens that ultimately earns its way into my "kit".
Optical Stabilization and "Hand-holdability"
I. INTRODUCTION
In this section I describe the results of head-to-head field-testing of the optical stabilization systems (and "hand-holdability") of the "new" Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR prime lenses. The goal of my testing was to discover how low of a shutter speed I could hand-hold each lens at and still get very sharp shots (and "keepers") using each of both lens's stabilization and/or customization settings designed for use on static subjects (including with the stabilization systems off). My preliminary testing was done while shooting 3-frame bursts on a Nikon D5. Follow-up - and more extensive - testing was done using multiple repetitions of longer 10-frame bursts more characteristic of how many wildlife photographers work in the field.
II. The THREE SENTENCE SUMMARY
There was extreme similarity in the shutter speeds at which I could hand-hold the Sigma Sport 500mm f4 and Nikkor 500mm f4E VR at and still obtain both very sharp shots and slightly less sharp "keepers" when shooting bursts of 3 shots. When I shot several longer (10 frame) bursts of shots using the various stabilization settings on the two lenses I did find some differences between the effectiveness of the settings and the lenses. I obtained a slightly higher number of sharp shots and overall number of keepers with the VR settings on the Nikkor 500 compared to the OS settings available to a Sigma 500 user without access to a USB dock and Sigma Optimization Pro software (i.e., when using the default OS "view" settings on the Sigma lens), but this difference disappeared when I used the Sigma lens with one of its OS customization settings (OS Moderate View).
III. The FULL EXECUTIVE SUMMARY
When it comes to the comparative effectiveness of the optical stabilization systems of the Nikkor 500mm f4E VR vs. the Sigma Sport f4 prime lenses in assisting a photographer when hand-holding the lenses the most accurate summary is this: Just the same, only different. Which means you can pretty much get to the same place (in terms of how slow a shutter speed you can hand-hold the two lenses at), but you have to take different pathways...and each of those pathways is quite different from the others. If you compare the two lenses as they come out of the factory you are likely to find that the Nikkor will allow hand-holding at slightly slower shutter speeds. And you'll find that the image appears much more stable as you look at it through the viewfinder. But both the appearance of the image through the viewfinder and the effectiveness of Sigma's OS system can be modified (and, most importantly, brought into virtual parity with the Nikkor lens) with customization of the OS settings using the optional USB Dock and Sigma Pro Optimization software. One additional result should draw the attention of anyone who likes to shoot in bursts (which includes most wildlife and sports photographers...so most users of these lenses!) - there is ONE VR setting on the Nikkor lens (VR NORMAL) where the image jumps around so significantly BETWEEN frames in a high-speed burst that many would consider it close to unusable in the field (especially compared to the uber-smooth VR Sport mode and all of Sigma's view modes).
IV. RELEVANT BACKGROUND INFO
A lens' ability to counteract camera shake through image stabilization - along with the correlated characteristic of how slow of a shutter speed you can hand-hold a lens at - can have a huge impact on its overall "usability". The importance of effective image stabilization varies considerably between users. For many wildlife photographers a lens that must ALWAYS be shot from a tripod will have a LOT less utility (and it will be used a lot less than if it could be successfully hand-held). But for other wildlife shooters image stabilization is a trivial feature - they may be roadside shooters that always have a tripod available or use such high shutter speeds that the quality of a lens's image stabilization is almost academic. For ME image stabilization is absolutely critical and can be even more important than absolute lens sharpness. Why? A number of reasons. First, while all serious wildlife shooters work in low light on at least a quasi-regular basis, I am in low light environments (like the Great Bear Rainforest) very regularly. Second, I am commonly shooting in places where a tripod can not be used - such as from a smallish Zodiac inflatable boat. Finally, I often am hiking significant distances to get to shooting locations and often don't want the weight of a tripod added to my load (or, alternately, I bring such a small tripod that an optical stabilization system is STILL needed to get sharp shots). For me (and I think a lot of other wildlife shooters) the quality of a lens's image stabilization system will largely dictate how close I can get to that lens's "theoretical" maximum image sharpness (i.e., how much of the theoretical sharpness is "realized" in a field setting). I can easily imagine scenarios where I would choose a slightly less sharp lens if it had a better optical stabilization system over a slightly sharper lens with an inferior optical stabilization system (because MORE of the sharpness would be realized with the lens with the better stabilization).
The point of this preamble? I can't tell any other shooter how important optical stabilization should be for them - that's something they can only decide for themselves. So it's up to you to decide the value of this part of my field-testing (and this section) to you. Could be critical...could be irrelevant.
Before going any further there are two other topics I have to go into a little detail about. First...what matters to me (and I think most shooters) isn't the absolute "measurement" of a lens's optical stabilization system (which, more often than not, is reported as the number of stops of camera shake and vibration that is "cancelled out" by the system). What matters to ME is the shutter speed that I can hand-hold the lens at and still get both tack sharp shots and "keepers". While this is correlated with the quality of the image stabilization system, other variables can be important. These other variables include lens/camera balance (very critical) and lens weight (sometimes critical...but its importance varies DRAMATICALLY between users). The results and findings reported here are primarily about lens "hand-holdability". And, in recognition of how we REALLY shoot in the field (in variable length BURSTS of shots)...the results you'll see below are largely expressed as proportion of sharp shots and keepers in BURSTS of shots.
Note that today's results are about the shutter speeds that I can hand-hold these two lenses at. You may be able to do far better (or a little worse) than me. So the absolute results are probably of little value to anyone. But there should be some value and generalizability in the comparative results - the shutter speeds I can hand-hold the Nikkor lens at VERSUS the shutter speeds I can hand-hold the Sigma lens at (i.e., which can I hand-hold at slower shutter speeds at?).
Second...for this entry to make any sense (and have any value) I have to go into some detail about the different modes and settings (and, for the Sigma lens, the customization available to the stabilization system) of the two lenses. There is definitely some apples-to-oranges things to consider. So...
A. Nikkor 500mm f4E Stabilization System
Nikon uses the term Vibration Reduction (or VR) to describe their system. It has 3 modes that can be selected via a toggle switch on the lens: VR OFF, VR NORMAL, VR SPORT. It is important to note the BOTH of the VR settings (Normal and Sport):
• Support panning AND stationary subjects, and
• both can be used while on a tripod. But just to confuse things Nikon...in their way...states the following their 500mm lens owner's manual:
"NORMAL and SPORT vibration reduction can reduce blur when the camera is mounted on a tripod. OFF may however produce better results in some cases depending on the type of tripod and on shooting conditions" (and, of course, the manual says NOTHING about WHAT shooting conditions they mean).
What's MY experience with the Nikkor 500 f4E VR system and tripods? Identical to what I have found on the 400mm f2.8E VR - if you are leaving the head loose so you can pivot the lens around (like most wildlife shooters do with gimbal heads)...just leave the VR on at most shutter speeds (I have YET to find any negative consequences on image quality of leaving the VR on at shutter speeds beyond which it is helping much, i.e., at 1/750s or faster the VR system is very likely adding little if anything to image sharpness, but isn't harming the shot). If you are tightening down the tripod head and shooting at very slow shutter speeds (like 1/60s or slower) - turn the VR off and do whatever you can to reduce vibration...including using a cable release, Live View, and Mirror Up settings.
OK...so what's the difference between VR NORMAL and VR SPORT settings? If you go to the manual and read their descriptions the gist is this: you get the MOST image stabilization with the VR NORMAL setting (they call it "enhanced" vibration reduction) and VR SPORT setting is best for photographing "athletes" and subjects that are moving rapidly and unpredictably.
What's MY experience with the different settings? I can't disagree with the comments in the manual, but they're kind of obtuse and quite lacking. In the real world here's what I've found: Yes, the VR NORMAL mode does give slightly better vibration reduction at REALLY low shutter speeds (see results below) and the VR SPORT mode is better for "action". But what's lacking is a basic difference between the modes that makes a HUGE difference if you're shooting ANYTHING in bursts (including static subjects): In VR NORMAL mode there is a huge amount of image movement (jumping) in the viewfinder BETWEEN frames in a high-speed burst while in SPORT mode the image is remarkably stable (like rock solid) between frames. This is especially noticeable when using a D5 (where the reduced image blackout time and new mirror-driving mechanism combines with the SPORT VR mode superbly). I've referred to the "image jumping between frames" characteristic before (when discussing the performance of the Nikkor 400mm f2.8E VR) as the "Herky-Jerky" (or HJ) factor. Note that the difference in this HJ factor between the two modes is so extreme that I simply will NOT shoot bursts with Nikkor 400mm f2.8E VR or the Nikkor 500mm f4E VR in VR NORMAL mode. And when a wildlife photographer says they won't use a particular feature when shooting bursts it pretty much means they refuse to use it at all.
B. Sigma Sport 500mm f4 Stabilization System
Sigma uses the term Optical Stabilization (or OS) to describe their system. Like the Nikkor 500 system it has 3 modes that can be selected via a toggle switch on the lens: OS OFF, OS1, OS2. And that's where the similarities end.
OS1 Mode cancels both vertical and horizontal movement/vibrations. This is the mode you use when hand-holding the lens with stationary or slowly moving objects. Basically...you use this mode for everything EXCEPT panning.
OS2 Mode cancels vertical vibration only...this is the mode you use if panning (it doesn't try to cancel out horizontal motion).
Tripod use? Sigma's lens manual says to NOT use either OS mode if your camera is mounted on a tripod. My own experience says you use the OS on a tripod just like you use Nikon's VR system on a tripod - if the tripod head is loose leave the OS system on and it will still provide image stabliization benefits. At slow shutter speeds (about 1/60s and longer) DEFINITELY turn it off - you can watch the image "drift" across the frame (a little) when the VR is on.
Customization of OS Settings? OK...now the plot gets thicker. If you own Sigma's USB dock and have Sigma Optimization Pro software you can modify the OS settings. There are three "view" settings that can be applied to either of the OS modes: Standard View (the default the lens is set with and the ONLY setting you have if you don't have the USB dock and Sigma software), Dynamic View, and Moderate View. Here's what each is supposed to do (and I am quoting the dialog boxes in the software):
• Standard View: "The OS effect is well-balanced and suitable for various scenes."
• Dynamic View: "This mode offers a recognizable OS effect to the image in the viewfinder. This helps to ensure the composition of images quickly."
• Moderate View: "This mode offers an excellent compensation of camera shake and achieves very smooth transition of the image in the viewfinder. The composition of the image remains natural even when the angle of view keeps changing."
Now if you can read these descriptions and figure out when you should use each setting you are a far better person than I am. And...if you can read these descriptions and conclusively decide between two possible interpretations - that the different "view" modes change the full operation of the OS (including the AMOUNT of stabilization) or they only change the view through the viewfinder - then you are ALSO a better person than I am! So...you guessed it...I took it upon myself to suss this out in my testing. One thing I CAN say about the descriptions - with the Dynamic View (and before shooting) you definitely see a more highly stabilized image through the viewfinder (much more like the stable view you get with the VR NORMAL view of the Nikon system). But...as you'll see...stable through the viewfinder doesn't necessarily mean "hand-holdable at slower shutter speeds".
One point I can't stress enough: If you do NOT have access to Sigma's USB Dock and their Sigma Optimization Pro software you have only ONE OS setting for each of the two modes (i.e., for OS1 and OS2 modes) - Standard View. With the USB Dock and the software you have two additional "customization" modes to choose from for OS1 and OS2 - Dynamic View and Moderate View.
V. WHAT I DID
Basically I shot a ton of hand-held shots with the two lenses mounted on my D5 in two different tests, both of which recognized the reality of how wildlife shooters actually shoot - in bursts. In the first test I shot 3-frame hand-held bursts of a large road sign at 40 meters. The sign was large enough to more than fill the viewfinder and image sensor. The sign has sharp edges on the lettering, multiple cracks varying in width, and a textured surface - all of which assist in making sharpness differences between images extremely easy to see. I shot 3-frame bursts at shutter speeds from 1/1600s down to 1/30s in 1/3 stop increments. Because BOTH VR modes on the Nikon lens tries to cancel both horizontal and vertical camera and lens shake - and thus are suitable for shooting static or very slowly moving subjects - I tested the Nikkor 500 f4E using all 3 VR modes (VR OFF, VR SPORT, and VR NORMAL). With the Sigma Sport 500mm f4 only the OS1 mode attempts to cancel both horizontal and vertical movement (and the OS2 mode cancels only vertical movement and is designed for panning), so I tested using OS1 mode only. However, if one has access to Sigma's USB Dock and Optimization Pro Software the OS1 setting can be set to three different custom "view modes" - Standard View (the lens default), Dynamic View, and Moderate View. Consequently I tested the Sigma lens using the following 4 modes: OS OFF, OS1 Standard View, OS1 Dynamic View, OS1 Moderate View.
The goal of this first round of testing? Nothing more than a coarse reconnaissance of the problem, to get a feel for how the systems performed, and to get a feel for the range of shutter speeds I had to examine in more detail to "tease apart" the differences in how the various VR and OS modes of the two lenses performed.
In the second round of testing I shot longer 10-frame bursts (which I believe are more representative of how a lot of wildlife and action shooters work in the field) of a yard torch at a distance of 10 meters. Like with the road sign there was a combination of sharp edges, textures, and surface cracking that made assessing sharpness differences between shots simple. For each VR or OS setting (as described above) I shot 10-frame bursts at shutter speeds from 1/500s to 1/30s in 1/3 stop increments using my D5. Between each "shutter speed run" (for a single OS or VR setting) I switched lenses (to mitigate against any form of bias associated with differential fatigue of my arms). I repeated the entire procedure (meaning the testing of all shutter speeds on all settings) four times.
The goal of this second round of testing? To reveal more subtle differences in the performance of the various VR/OS settings (and to let ME know what shutter speeds I could use at specific shutter speeds and what settings I preferred).
Scrutinizing the images: I assessed image sharpness via examining previews of the raw images using both Lightroom CC and Capture One Pro 10. Images were viewed at 100% magnification (1:1) on a 30" Apple Cinema Display with a native resolution of 100 pixels per inch (ppi). I chose this monitor for all image comparisons because small differences in sharpness are often "masked" when images are viewed at 100% on some newer high resolution displays (e.g., almost ANY image looks sharp on my 218 ppi iMac 5K Retina Display). Yes, like with my testing on optical performance I was basically pixel-peeping!
I categorized all images into one of 6 sharpness categories (as per my 20 Nov 2015 assessment of images shot when comparing 3 "user-zooms" - see that entry right here...). The categories were:
• Sharp: All detail on central portion of target object absolutely sharp (good 'ol "tack sharp")
• Slightly Soft: Shows any softening of detail in central portion of sign, but careful sharpening in Photoshop would make these shots indistinguishable from those categorized as "Sharp". Still a "Keeper".
• Softer: Noticeable softening of image detail; sharpness loss NOT fully recoverable by digital sharpening. NOT a "Keeper".
• Soft: Unacceptable image sharpness
• Very Soft: Close to garbage!
• Very, Very Soft: Pure garbage, AKA a bloody and blurry mess!
While subjective, in practice I had NO problem quickly assigning an image to one of these 6 categories. Only images in the first and second categories (Sharp and Slightly Soft) are images I considered (and later classified) as "Keepers".
VI. RESULTS
The KEY results have been described above in the THREE SENTENCE SUMMARY and the FULL EXECUTIVE SUMMARY. Those wishing to see more detailed results can view them in Appendix 2 (below).
VII. DISCUSSION AND CONTEXT
Here's some of my own thoughts about what my testing of the optical stabilization system of these two high-end super-telephotos really mean. First, there's simply no doubt the systems work and allow the user to shoot hand-held shots at lower shutter speeds than possible without the stabilization. No matter how I slice and dice the results I really can't find a 4-stop improvement in the shutter speeds I can hand-hold either lens at (as Nikon claims in their marketing literature), but it is important to note that I am only indirectly testing the stabilization systems (my results are all about "hand-held" shutter speeds and those shutter speeds depend on variables beyond JUST the stabilization system) and the metric Nikon uses to come to that 4-stop claim may be valid in the test they use. But it didn't translate into a 4-stop advantage at shutter speeds I could use for hand-holding their 500mm lens.
Second, if you compare completely "stock" versions of the two lenses (before customizing the Sigma lens) you'll likely get slightly better image stabilization performance out of the Nikon lens (compare the shutter speeds above with both VR SPORT and VR NORMAL modes of the Nikon lens to the DEFAULT OS1 mode of the Sigma lens - i.e., to OS1 Standard View). If you purchase the optional USB Dock and use the Sigma Optimization Pro software to customize the OS1 mode of the Sigma lens you can get virtually identical stabilization results with the Sigma lens as you can with the Nikon lens (compare the OS1 Moderate View results above with the VR NORMAL and VR SPORT results). So...those who aren't into the "technical end" of things - and just want to pick up their lenses at the dealer and just want to shoot with them without taking the time to customize and set them up would probably be better off with the Nikkor 500.
Third, if having a stable image when looking through viewfinder is an important part of a stabilization system to you then you should either select the Nikon lens OR be prepared to customize the Sigma lens (in this case to OS1 Dynamic View). The stability of the view through the viewfinder is VERY similar with Sigma's OS1 Dynamic View and Nikon's VR SPORT and VR NORMAL mode.
Fourth, if you want stable images through the viewfinder DURING a high-speed burst then use ANY mode you want to other than the VR NORMAL mode on the Nikkor 500mm lens!
My favourite modes on each of the lenses? On the Nikkor 500 - it's absolutely the VR SPORT mode (and this is the mode my Nikkor 400mm f2.8E VR is always set to). On the Sigma Sport 500? It depends. I LIKE the stability in the viewfinder of the OS1 Dynamic View mode, but it doesn't offer quite as much stabilization as the OS1 Moderate View mode. SO...if I'm in situations where I'm hand-holding the lens at shutter speeds no slower than about 1/400s I prefer using the OS1 mode with Dynamic View. BUT...if the light drops and I need to use slower shutter speeds I prefer using OS1 Mode with Moderate View. And, I have set the lens up so that the C1 setting on the customization switch puts me in Dynamic View mode and C2 puts me in Moderate View mode - so it's EZ-PZ (and fast) to shift between the view modes in the field.
My OWN take home lesson when hand-holding these two lenses in the field? I can be quite confident that I will obtain a high rate of sharp shots and a very high rate of keepers if I shoot either lens at 1/250s or faster. At slower shutter speeds my rate of sharp shots and keepers will begin to fall, and it would be prudent to shoot either slightly longer bursts, more bursts, or both!
See...when it come to the optical stabilization of these two lenses they're just the same, only different! ;-)
Autofocus Performance
I. INTRODUCTION
In this section I describe the results of head-to-head field-testing of the autofocus (or AF) performance of the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR prime lenses. I compared the AF performance of the two lenses in three ways. The first way is purely subjective: Did I notice any functional difference in AF performance between the lenses while "just shooting" the two lenses. So things like "Did one lens seem snappier in acquiring initial focus?" or "Was one of the lenses noticeably faster in moving from closest focus to distant focus?" or "Did I notice one lens struggling more than the other to focus on certain types of subjects, such as those in low-light or heavy shade?". This type of autofocus comparison isn't at all quantitative but most experienced photographers know that what they notice while "just shooting" DOES matter to them (and, if it's a valid observation, often shows up in more rigorous testing).
The second thing I looked at was a comparison of both "keeper rates" and the proportion of very sharp shots (vs. slightly softer shots vs. out-of-focus shots) when I repeatedly shot high-frame rate sequences of a rapidly moving subject. This test (explained in more detail below) combines aspects of a lens' AF speed and both the predictive AF capabilities AND the tracking abilities of the lens/camera system in use. In this test I use a "proxy" of wildlife (namely, my Portuguese Water Dog Poncho) that is far more cooperative than any truly wild animal and allows me to replicate the same test many times (limited only by the amount of treats my pocket can hold!).
The final thing I looked at was the accuracy of the phase detect (i.e., through the optical viewfinder) AF system over the entire array of selectable focus points on both the Nikon D5 and D500. In this test I compared the sharpness of my subject that was focused on using Live View (contrast detect) focusing vs. its sharpness when focused on using the "traditional" viewfinder-based (phase detect) focusing. In essence, all I was really trying to answer with this question was this: Are both lenses capable of focusing accurately on all focus points in the focusing array, including the outermost focus points (that come close to touching the edge of the frame) on the Nikon D500?
II. BACKGROUND
Autofocus performance is one of the most critical components of lens performance for most users of super-telephoto lenses. Both sports and wildlife photographers regularly use these lenses to capture moving subjects - in the case of wildlife photographers it's often those good ol' BIF (Bird In Flight) shots or even shots of running mammals (things like this shot of a grizzly running in water).
A fairly prevalent belief among many wildlife shooters (and one I've heard many times when chatting with other wildlife photographers) is that while some third party lenses may challenge or even match the "best of the best" lenses from Nikon and Canon from an optical perspective, they almost always fall behind the lenses from Nikon and Canon in AF performance (at least a little). One of the most commonly cited reasons for this is that because 3rd party lens makers like Sigma or Tamron are competing with Nikon and Canon for photographers' lens-buying dollars the camera makers aren't likely to share detailed proprietary information on the engineering and technologies of the AF systems of their own lenses and bodies. Thus, third party lens makers are forced to "reverse engineer" the AF systems of Nikon and Canon and, logically, they tend to lag a bit behind Nikon and Canon in AF technology. I admit that until I field-tested and compared the AF performance of the Sigma Sport 150-600mm against its closest competing zoom lenses from Nikon (the AF-S 80-400mm and the AF-S 200-500mm) in 2015 I had the same bias (i.e., the belief that the AF performance of the third party lenses would not match that of Nikon's own lenses). That testing showed that the AF performance of the Sigma Sport 150-600mm was actually slightly better than that of the 80-400 and the 200-500 (but not as good as Nikon's best primes). Those tests are reported in my 2015 blog (see blog entries of 25 March 2015 and 03 April 2015 for comparisons of AF performance of the Nikkor 80-400 vs. the Sigma Sport 150-600, and blog entry of 03 Nov 2015 for comparison of AF performance of both the Nikkor 80-400 and the Nikkor 200-500 vs. the Sigma Sport 150-600).
It's important to note that the Sigma Sport 500mm f4 lens features a brand new AF motor (i.e., a new Hyper Sonic Motor) that Sigma claims has 1.3x the torque of previous models which, according to Sigma, translates into "...fast and snappy focusing." How much faster and how much snappier is unclear.
So the main question I'm left with (and have been wondering about ever since I decided to do this field-test) is one that a LOT of shooters probably have: With Sigma's recent commitment to producing world class lenses, and with that new AF motor on the Sigma 500mm, have we finally reached the point where Sigma and other third party lens makers can rival the "big guns" in AF performance?
An Important Note to Canon Users: While many of the findings of this "500mm Wars" series probably have a high degree of applicability to the Canon version of the Sigma Sport 500mm f4 (e.g., there is no real reason to expect that the Canon version will differ from the Nikon version in sharpness or OS performance) I would not assume that the Nikon and Canon versions of the Sigma 500 are identical in AF performance. It is possible that if you mounted a Sigma 500 on a Nikon D5 and on a Canon 1DxII they would exhibit very similar AF performance, but that absolutely can NOT be concluded from my findings - I used Nikon bodies only. And AF performance is a joint effort between camera body and lens.
III. The KEY THREE SENTENCE SUMMARY
The differences in AF performance of the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR are so subtle that they are unlikely to be noticed under normal day-to-day shooting; both lenses show a very snappy initial focus acquisition, both shift from closest focus to distant focus very fast, and both re-focus quickly and smoothly enough that they rarely miss focus on even fast moving subjects. Repeated trials of continuous high-frame rate shooting on a rapidly moving subject showed "keeper rates" of almost 90% for both lenses, but with the Sigma Sport having a slightly higher rate of sharp shots. Both lenses exhibited high focus accuracy on all 55 selectable focus points of a Nikon D5, but the Sigma Sport 500mm f4 showed a higher degree of focus accuracy on several of the outermost focus points on a Nikon D500.
IV. The FULL EXECUTIVE SUMMARY
In day-to-day field use and when "just shooting" it is extremely difficult to detect any real-world difference in the AF performance of the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR. Both are very fast at acquiring initial focus. Both can quickly refocus on subjects at very different distance (e.g., from only meters away to kilometers away). Both are very good at acquiring and holding focus on moving subjects. In short, both have excellent AF systems.
After running repeated trials where I shot continuous high-frame rate sequences of a rapidly moving subject designed to mimic a running wild animal I was able to detect only very small differences in keeper rates and the proportion of sharp shots that either lens could obtain. When these trials were performed using 72-point dynamic area focusing on a D5 the Sigma Sport 500mm f4 produced an overall "keeper" rate of 90.2% of the shots, with a full 50.5% of the shots being judged as very sharp. When using 72-point dynamic area focusing on the D5 with the Nikkor 500mm f4E I obtained a keeper rate of 87.1%, with 41.8% of the shots being very sharp. When I switched to 9-point dynamic area focusing I found the same overall trend, with the Sigma 500 very slightly out-performing the Nikkor 500 in both keeper rates and percentage of sharp shots. In these 9-point trials the Sigma 500 produced a keeper rate of 88.7% with 49.3% of the shots being very sharp. In 9-point dynamic area mode the Nikkor 500 produced keeper rates of 84.9% with 36.0% of the shots categorized as very sharp.
An examination of focus accuracy of the viewfinder-based AF system (i.e., the phase detect system) across the array of focus points also showed only small differences between the two lenses, again with a slight advantage going to the Sigma 500. When using a full-frame sensor D5 camera both 500mm lenses performed excellent and equally - I could find no difference in focus accuracy between Live View (contrast detect) and viewfinder-based (phase detect) focusing on any of the 55 selectable focus points. However, when I did the same comparison using a DX (cropped) sensor D500 (where the outermost selectable focus points reach closer to the edge of the viewfinder) there was a difference in focus accuracy between the two lenses: With the Sigma 500 there was no difference in focus accuracy (between the Live View and viewfinder-based) in any of the 55 selectable focus points, whereas with the Nikkor 500 there was no difference in focus accuracy of 51 of 55 of the selectable focus points. The four points that the viewfinder-based AF system could not accurately focus on when the Nikkor 500 was in use were the extreme corners (left-most top point, left-most bottom point, right-most top point, right-most lower point).
V. MORE DETAILS:
For a LOT more information about the nitty-gritty details of what I did and what I found in assessing AF performance please jump down to Appendix 3. It does contain some observations that go slightly beyond what is described above (including comments about one idiosyncrasy found in AF performance of the Sigma 500 during some high-speed bursts) in the Executive Summary, but those with little time to spare can probably quit reading now! ;-)
VI. CONTEXT...AND A FEW FINAL THOUGHTS ON AF PERFORMANCE
OK...where does this leave us with respect to autofocus performance of these two great lenses? In summary, I found that the Sigma 500 has ONE autofocus idiosyncrasy (that slight focus shift between frames that's noticeable in SOME high-speed bursts - details available below in Appendix 3) that may bother some. But, the Sigma 500 very slightly outperformed the Nikkor 500 when it came to holding focus on a rapidly moving subject and was very slightly better in focus accuracy on 4 of 55 selectable focus points on a Nikon D500 (but NOT on a Nikon D5). In my mind - and collectively - these are incredibly trivial differences that would almost never have a significant (or noticeable) impact in a field setting. Before I began this test I fully expected that the Nikkor would edge out the Sigma in AF performance. Now, I'm of the feeling that they are virtual autofocus "clones" of on another - and I really wouldn't "gravitate" towards one lens or the other based on the trivial differences I found in AF performance. It is possible that with continued use I (or someone else) will discover some other difference in AF performance between these two lenses. But I doubt it will be a major or significant difference.
Musings From Just Shootin'
I've mentioned several times in this field test that when I test a lens I combine fairly rigid and systematic field-based tests with sessions where I just SHOOT with the lens(es) in the way I would when "at work". The time spent "just shooting" under less controlled conditions serves multiple purposes, including demonstrating to me how closely the daily use of the lens will match the "theoretical best" performance it exhibits under highly controlled conditions. This is critical to ME in that I almost never shoot under controlled conditions in the field - as a wildlife photographer who works only with free-ranging, non-constrained, and fully wild subjects I end up doing a whole of "cowboy shooting"! Additionally, the time spent "just shooting" gives me a feel for how several different and independent parameters of lens performance interact in the field. For instance, when hand-holding a big super-telephoto lens in the field at least 3 factors can impact on the quality of the resulting images - the lens weight and balance, the optical stabilization system, and the autofocus system. At times we test or evaluate these factors independently, but the reality in the field is that they interact in producing that final image that we either keep or throw away.
So...here's some musings about things I've noticed about how the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR have performed in the field...
I. PHYSICAL CHARACTERISTICS - IN THE FIELD
As noted in the first section of this review, there are far more physical similarities between these two lenses than there are differences. In my view only three of the differences between them are likely to have any significance to most users.
A. Lens Weight:
While both of these lenses are significantly lighter than Nikon's "old" 500mm f4G super-telephoto, the Nikkor 500mm f4E is 324 gm (or 0.71 lb) lighter than the Sigma Sport 500mm f4.
The sixty-four thousand dollar question: Is this difference noticeable in the field? Sort of. What I mean by this is that during times when I was doing either formal lens testing or when "just shooting" (involving hand-holding of the two lenses) where I rapidly swapped between the lenses I DID notice the weight difference. But, if I was throwing one of the lenses in a backpack OR walking around in the field with one of the two lenses in my hand (using the tripod foot as a handle), I couldn't have told you by weight alone which lens I was carrying. And, more importantly for me, even with the optical stabilization systems turned off, I found there was virtually no difference in the shutter speeds at which I could effectively hand-hold the two lenses at (I've long thought that balance of the lens-camera system is more important in "hand-holdability" than absolute weight is).
So, for me, the difference in these two lenses in weight is quite inconsequential. However, I would not say this will be the case for everyone. I have no doubt there will be some shooters out there that find they can carry or hand-hold the Nikkor 500mm f4E VR more easily than the Sigma Sport 500mm f4. And, of course, there will be some shooters who struggle to hand-hold both of these lenses, and some who could easily carry or hand-hold the lenses if they were twice the weight!
B. AF Function (AF Activation) Buttons:
This might be a small thing to a lot of users, but it isn't for me. In recent years all Nikkor super-telephoto lenses have come with four buttons arranged in a ring around the distal end of the lens. The concept is that the buttons could be used to control one of several aspects of the lens function (e.g., AF-On or AF-lock) with your left hand (that is positioned near the end of the lens) while you were shooting. Cool idea. The Sigma Sport ALSO has these 4 buttons. But rather than having them on the exact top, bottom, and either side of the lens (the "cardinal" positions - so positioned at 90º, 180º, 270º, and 360º) the buttons are slightly rotated (to the left) and offset from the cardinal positions. In my own case this means the buttons on the Sigma 500 fall in a much better "natural" positions for my thumb (as in, directly below my thumb), especially when I'm hand-holding the lens.
When I'm shooting in the field I use the lens AF Function/Activation buttons to switch between autofocus area modes (this feature is NOT available on all Nikon DSLR's). For example, my preferred "default" AF area mode on the D5 is 9-point Dynamic Area mode, but when using the Nikkor 500 or the Sigma 500 I can switch to a different area mode by pressing and holding the AF Function/Activation button on the lens (I normally use Group Area as my alternate AF area mode, but sometimes I change this in the field).
So...how important is the positioning of the AF Function/Activation buttons? Well...I LOVE the "offset" of the Sigma buttons (relative to the Nikkor buttons) and find them MUCH more usable. Consequently I am using the AF Function button on the Sigma Sport 500 much more often than I have ever used the AF Activation buttons on any of the Nikkor super-telephotos I have owned.
So - for ME - consider this to be "...a BIG little thing". Advantage Sigma.
For other users? I'm sure there are some users of Nikkor super-telephotos that never use these buttons. That MIGHT be due to their arrangement/position. Or, it might be that the user just couldn't be bothered. And, I am sure that there will be some users who love the Nikkor AF Activation buttons and some of those may react exactly as I have to Sigma's re-positioning of these buttons. So the importance of this between-lens difference in positioning of the AF Function/Activation buttons will likely vary tremendously from one shooter to the next.
C. Sigma's USB Dock and Lens Software Configurability:
This is a tough topic to pigeonhole - you normally don't customize your lens while still in the field, but it directly IMPACTS on how the lens performs in the field. In terms of background - most recent lenses from Sigma are customizable, but only if you purchase their USB dock and download and install their free Sigma Optimization Pro software. In the case of the Sigma Sport 500mm f4 lens aspects of both autofocus and optical stabilization can be customized to those with the USB dock and software. And, of course, the software and firmware of the lens can be upgraded by the end-user (as Sigma makes updates available).
One major value of Sigma's system really hit home for me when Nikon announced the need to fix the firmware on their 200-500mm f5.6 VR zoom. A free fix, of course, but you had to box up and ship your lens to Nikon to have them fix the bug. I live in a relatively remote location where couriers won't even come close to my home (it's a 50 km drive to get to the closest FedEx or Purolator depot)...and mail service is slow. So for me the "free" fix from Nikon on their 200-500 would have meant being without the lens for a minimum of one week, and realistically closer to two weeks. Same problem with the Sigma would take me about 5 minutes to "fix" using their USB dock and their Sigma Optimization Pro software.
Ok...but what about the "customizability" of the AF and OS functions? Does choosing different settings really make much difference in the field? Yes, a lot. I will go so far as to say that WITHOUT customization (i.e., using the default lens AF and OS settings the lens comes with and that users who DON'T get the dock will be forced to live with) - and for MY USES of the competing lenses - the Nikkor 500mm f4E VR outperforms the Sigma Sport 500 f4. And, more importantly, I would select the Nikkor 500mm f4E OVER the Sigma Sport 500 f4, even given the huge price difference. Said another way, with a little effort and time (and if you buy the USB dock and Sigma Optimization Pro software), you can fully nullify the performance differences between the lenses by "tweaking" the AF and OS settings on the Sigma lens. In my books, that's HUGE (or is it YUGE?).
And...that leaves with me a few thoughts...
• I think the customizability of the Sigma 500 is so essential to its performance that the USB dock should come WITH the lens (i.e., included with the lens purchase), even if it means the price of the lens has to go up by $50 or so.
• After extensive testing of both the OS system of this lens and the AF performance of this lens (still to be reported in detail), and after considering the target market for this lens (sports and wildlife/nature photographers) I think Sigma has selected the wrong default values for the customizable functions on the lens. Given the USB dock is an optional accessory (that many may choose NOT to buy), this may lead to a large number of owners to be less happy with the lens than they could be. And it may even lead to some lens reviews that might turn out to be less positive than they could be.
It's also worth noting that a lot of folks looking at buying this lens (or any lens) just want to put the lens on their camera and USE IT. They don't want to fine-tune the focusing, and they don't want to take and examine thousands of test images to separate out the nuances of the different custom settings. So "giving them" default custom settings that are a little more carefully chosen (to match the intended user) is kind of important.
II. OPTICAL PERFORMANCE - IN THE FIELD
The outcome of my "controlled" testing of the optical performance of the Sigma and Nikkor 500's was almost astonishing to me - while I didn't expect to find major differences in optical performance, I honestly did NOT expect the two lenses to be virtual clones of one another optically. But what happens to that optical performance when you wander into the field and lose a lot of your control of the shooting situation?
OK...one caveat here: I'm not too into "roadside" wildlife photography and I often hike significant distances to locate my "prey". Which can make doing head-to-head field comparisons of wildlife images shot with "dueling 500's" pretty challenging (carrying ONE 500mm lens into the woods is often challenging in itself...carrying two is a complete pain in the butt!). But, nonetheless, I have found ways to work with the two lenses with wildlife (and my not-so-wild Portuguese Water Dog Poncho) and can say this: When shooting under real world field conditions the optical parity of these two lenses IS retained. This is huge.
So...here's a FEW sample image pairs to look at. I do have many more images to come in my final review. And it's important to note that all image pairs below share absolutely identical camera settings. All were captured as RAW files and processed IDENTICALLY using Phase One's Capture One Pro combined with Adobe Photoshop CC 2017. All Nikkor 500mm f4E shots were captured using VR Sport mode. All Sigma Sport 500mm f4 shots were captured using OS1 optical stabilization mode (customized to Moderate View mode).
A. Bighorn Ram on Ridge:
Image notes: Nikon D5. 1/500s @ f6.3 and ISO 250. Supported on Really Right Stuff TVC-24 tripod with Jobu Jr. 3 Deluxe gimbal head (left loose).
• Download Sigma 500mm f4 Image (2400 pixels; JPEG: 1.42 MB)
• Download Nikkor 500mm f4E Image (2400 pixels; JPEG: 1.42 MB)
B. Bighorn Lamb and Rabbitbush:
Image notes: Image notes: Nikon D5. 1/500s @ f6.3 and ISO 1250. Supported on Really Right Stuff TVC-24 tripod with Jobu Jr. 3 Deluxe gimbal head (left loose).
• Download Sigma 500mm f4 Image (2400 pixels; JPEG: 1.7 MB)
• Download Nikkor 500mm f4E Image (2400 pixels; JPEG: 1.7 MB)
C. The Sprint - Poncho the Portie:
Image notes: Nikon D5. 1/2000s @ f5.6 and ISO 320. +0.67 stop compensation from matrix-metered exposure setting. Hand-held.
• Download Sigma 500mm f4 Image (2400 pixels; JPEG: 1.0 MB)
• Download Nikkor 500mm f4E Image (2400 pixels; JPEG: 1.0 MB)
III. OPTICAL STABILIZATION AND HAND-HOLDABILITY - IN THE FIELD
The conclusion of my "controlled" testing of the hand-holdability of the two 500's showed that I could hand-hold the two lenses down to the same range of shutter speeds when shooting in bursts.
But when I'm shooting in the field I generally notice - and place importance on - three aspects of optical stabilization.
A. Slowest "Hand-holdable" Shutter Speed
Like all wildlife shooters I am continually looking for the most suitable balance of shutter speed, aperture and ISO. I often work in low light environments and, over the years, have developed a tendency to try to keep to shutter speeds no slower than 1/focal length of the lens when hand-holding super-telephoto lenses (so 1/500s with a 500mm lens). I try to stick to this, but if that pushes the ISO over the "threshold" of the camera in use, then I will let shutter speeds drop further...sometimes to 1/200s or even slower.
What am I finding with the Sigma 500 and the Nikkor 500 in the field? Pretty much what I found in controlled tested - that "in the right mode" both lenses let me shoot freely at 1/500s and get an incredibly high rate of tack sharp hand-held shots (almost 100%). And, both lenses let me go down to 1/250s and still get a very high keeper rate (way over 50%). If I go to crazy slow shutter speeds (where you start risking having a shot ruined by SUBJECT movement) like 1/80s I'm still likely to get some sharp shots and several keepers in a short-to-moderate length burst. So the optical stabilization systems on BOTH lenses deliver well in the field.
What am I considering the "right mode" to use for the two lenses (i.e., what works best for me in terms of cancelling out camera shake)? With the Nikkor lens it can be EITHER of their two modes - Sport or Normal - right down to the "crazy slow" shutter speeds (like 1/80s). With the Sigma you virtually always use OS1 mode unless panning, but still have 3 custom view modes to choose between (assuming you have access to a USB Dock and Sigma Optimization Pro software) - Standard View (the default setting of the lens), Dynamic View, and Moderate View. If I have enough light to be hand-holding the lens in the 1/400s to 1/500s range I found that any of the 3 view modes can be used. Once shutter speeds start dropping down I invariably get the best results using Moderate View mode.
This hand-held shot of young Bighorn Sheep Ram taken with the Sigma 500 (at 1/500s and f7.1) typifies the type of results I am getting with Sigma Sport AND the Nikkor lenses in the field:
• Download Bighorn Ram Image (2400 pixels; JPEG: 1.5 MB)
B. Image Stability Through the Viewfinder (BEFORE Shooting)
Another thing that I notice in the field and DOES matter to me is how stable the image appears through the viewfinder BEFORE I start shooting using the various VR or OS modes. Why does this matter? For at least two reasons. First, it can impact on how well you can compose the scene while looking through the viewfinder, especially if hand-holding the lens. If you turn the VR/OS system of a super-telephoto off and try to careful compose a scene while hand-holding the lens you'll know what I mean. Second, if you need precise positioning of your focus bracket to get the shot you want it is MUCH easier if the image you are looking at is stable! It should also be noted that the stability of an image through the viewfinder does have a bit of a psychological component as well - if it APPEARS stable through the viewfinder then the user KNOWS the optical stabilization system is working and tends to be more confident in it (even if this confidence is misplaced...more about this below).
Note that I'm not alone in liking a stable image as seen through the viewfinder - I have received emails from people who have told me outright that they have purchased - and then subsequently sold - lenses that have stabilizations systems that didn't produce "stable enough" views through the viewfinder.
Anyway...what have I been finding in the field re: viewfinder stability of the two 500's in their various modes? The Nikkor is simple - BOTH VR modes (Normal and Sport) produce VERY stable images through the viewfinder. The Normal mode does produce a slightly more stable image through the viewfinder, but the difference between that and the Sport mode is almost negligible (luckily for Nikon...you'll see what I mean by the end of this section).
With the Sigma lens the three custom view modes are VERY different in viewfinder stability - and it's REALLY noticeable in the field. The Standard View (the default view mode on the lens, and the ONLY view mode available to the user if they don't have the USB Dock and customization software) provides very little stabilization of the image as you're looking through the viewfinder. When using this mode when hand-holding the Sigma 500 I actually checked several times to see if the OS system was still on (and I hadn't accidentally turned it off). The Dynamic View mode is at the other end of the "stability through the viewfinder" extreme - it produces the MOST stable image through the viewfinder. It's VERY equivalent to the Nikon Sport Mode in viewfinder stability. And the Moderate View? About halfway between the two other custom view modes - with "decent" stability of the image through the viewfinder (and in most cases it is stable enough to allow pinpoint placement of your focus bracket on a small portion of the subject).
One point I have to make here: One might expect that the stability of the image through the viewfinder is directly correlated with how slow of a shutter speed you can hand-hold the lens at (i.e., both characteristics are demonstrating the same thing - the degree of image stabilization). But, that doesn't appear to be true. I found that the Dynamic View of the Sigma lens (the view mode that provides the most image stability through the viewfinder) didn't allow me to hand-hold the Sigma 500 at as slow of shutter speeds as either the Standard or Moderate Views did!
C. Between-Frame Image Stability During Bursts of Shots
Last - but certainly not least - is something I discovered with my Nikkor 400mm f2.8E and also shows up on the Nikkor 500mm f4E: There can be a dramatic difference in how stable an image is BETWEEN frames in a burst depending on the VR mode you choose. Nikon has simultaneously achieved a new high AND a new low in performance here! Use your Nikon D5 with the 500mm f4E lens in Sport Mode when shooting a high speed burst and you'll be stunned how stable the image is throughout the entire burst (the camera body DOES make a difference here...the image is still quite stable when doing a high speed burst with a D500, but NOT as stable as when using a D5). Examine the images after the fact and you'll notice that EVERY shot in the burst is framed virtually identically - there is simply NO between-frame jumping in Sport Mode.
What happens when you switch to VR Normal mode? Well...just don't shoot a high speed burst in Normal mode if you're prone to motion sickness. You'll puke. It's that bad. And, if you look at the resulting images afterwards you'll learn that it's NOT just viewfinder behavior - the image WAS jumping around that much. In my view, if you shoot in bursts (and what wildlife photographer doesn't?) you really have only one usable VR mode on the Nikkor 500 - VR Sport.
What about the Sigma 500? Well, interestingly...there's almost no difference between the 3 custom view modes in between-frame image stability when shooting bursts. And, all three are very good and approach the between-frame stability of the Nikkor 500 in Sport Mode.
What are the take home lessons on the two 500's optical stabilization systems? Well...if you're the kind of wildlife photographer who wants to pick up a lens at the store and just put it on your camera and NEVER think about its VR settings again - get the Nikkor and put it in VR Sport Mode. If you're the type of user who is willing to experiment with your lens (including modifying customization settings) you can opt for the Sigma Sport and adjust it to perform almost identically to the Nikkor 500 in Sport Mode.
My "default" optical stabilization settings when I'm hand-holding the two lenses? For the Nikkor 500mm f4E VR it's VR Sport Mode. For the Sigma Sport 500 it's OS1 mode with Moderate View custom setting. And after a TON of testing and shooting I think Moderate View mode should be the default mode of the Sigma 500 as it comes out of the factory.
The FINAL Wrap-up and My Lens Choice
My primary goal in doing this extensive field test comparing the Nikkor 500mm f4E VR and the Sigma 500mm f4 Sport was to determine which of the two lenses I would select for my own wildlife kit. Ultimately this means I am seeking the answer to the following question: "Which of the two lenses will offer the best possible combination of performance and usability in the type of field situations I am likely to put it to?"
My secondary goal in doing this test was to answer an important question I have about the best combination of lenses for my work, specifically this: "Given I already own Nikon's superb 400mm f2.8E VR super-telephoto, do I really need ANY 500mm lens in my wildlife kit?". Those shooters who do own the 400mm f2.8E and are considering adding a 500mm to their stable of lenses should pay particular attention to the short section below on optical performance.
My own bias - and emphasis - in all my field testing is to focus on features and performance in the field (as I will use the lens). In the case of 500mm super-telephoto lenses this means I put extra emphasis on things like realizable image quality (where "realizable" = what I can obtain in the field), hand-holdability (which automatically factors in numerous variables, including lens weight, lens balance, optical stabilization, and more), and autofocus performance. In my field testing and reviews I am likely to ignore - and not mention - things like statements on dxomark.com telling me that the Sigma Sport 500mm f4 performs best on a 46 MP Nikon D820 camera (when I KNOW my copy of this lens will be used primarily on a Nikon D5 and Nikon D500). It should be clear - given there has been no announcement by Nikon of a 46 MP D820 - that this is just a theoretical example...but you get the point! I fully acknowledge that other wildlife photographers and, more likely, photographers from other genres of photography may choose to weigh the various lens performance variables differently than I do.
Finally, I did NOT factor in lens price or "dollar value" in my comparison of these two lenses. Price is something that is incredibly important to the "average" purchasers of camera gear but, even so, still varies significantly in importance between users (and for some buyers factors like brand loyalty over-ride it). And, of course, the amount of money one is prepared to spend on a camera lens is a personal decision.
I. THE FINAL SUMMARY
What follows are the absolute most significant findings of my 3+ months of testing and comparing the Nikkor 500mm f4e and the Sigma Sport 500mm f4 super-telephoto lenses. You'll see a few references to comparisons with other lenses (such as the Nikkor 400mm f2.8E) but the major focus is on the two 500's.
A. Physical Characteristics
While there are cosmetic differences between the Nikkor and Sigma 500, both of these Japanese-made lenses have excellent build quality. Both are environmentally sealed. The Sigma 500 is very slightly shorter than the Nikkor (both with hood off/reversed and slightly more so with the hood mounted) and this difference MIGHT make a difference for a small number of users (where the Sigma might fit more easily into their chosen carrying pack).
BUT...the most significant physical difference between these two 500's is in overall weight. In fact, it's probably the biggest single difference between the two lenses overall! BOTH are light lenses for super-telephotos, but the Nikkor IS a full 324 gm (or 0.71 lb) lighter. But it's important to have context here - the Nikkor 500 is one of the lightest super-telephotos ever made. The Sigma 500 is still slightly over a pound (454 gm) lighter than the OLD Nikkor 500 (i.e., the Nikkor 500mm f4G) and 1.25 lb (570 gm) lighter than the NEW 400mm f2.8E VR.
How important is the weight difference? For some - absolutely critical. For others - minimally important. There WILL be some users out there who can hand-hold the Nikkor 500 but not the Sigma 500. If someone falls into that category they would DEFINITELY be better off with the Nikkor 500mm f4E VR. In my case, I notice the difference in weight between the two lenses if I rapidly switch back and forth between them, but if I take just one of them into the field at a time I couldn't tell you which lens is in my hand (as I am carrying but the tripod foot) by weight alone. And...as you'll see below...for me the weight difference doesn't translate into how slow of a shutter speed I can hand-hold one lens at versus the other.
B. Optical Performance
I tested the comparative optical performance of the two 500's using both rigidly controlled systematic field-based tests and extensive periods of "just shooting". Here's the simple net result in a two sentence summary (pulled directly from the Optical Performance section above):
I have NEVER tested any two competing lenses that are so absolutely similar in image quality (at all distances, apertures, and with or without teleconverters) than the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR. Image sharpness, quality of the out-of-focus zones, and the progression in increasing sharpness from wide open through to about f5 (where both lenses approach maximum sharpness) is virtually identical between my copies of these two lenses.
I also compared the image quality of images shot with the Nikkor 400mm f2.8E AND the Sigma Sport 150-600mm (@ 500mm) to the shots captured with the two 500's. I found that the quality of images shot with the 400mm f2.8E and then UPSIZED (digitally increased the resolution of) did NOT match the quality of the images shot with either of the two 500's. I found that quality of the images shot with the 400mm f2.8E PLUS TC-14EIII (1.4x) teleconverter and then DOWNSIZED (or down-sampled) to 500mm "size" was very similar to the quality of the images shot with the two 500's (i.e., no significant or noticeable differences between the image quality) at wide apertures but when stopped down the two 500's did produce sharper images. And...while the Sigma Sport 150-600mm (@ 500mm) did produce shots with good image quality, the images shot with the two 500's WERE sharper and with smoother out-of-focus zones.
C. Optical Stabilization and "Hand-holdability"
We all know were SUPPOSED to put super-telephotos on a tripod - right?. But...in the real world there are times when we are simply forced to hand-hold them (like on all my Grizzlies of the Khutzeymateen Photo tours!). I tested the "hand-holdability" of the Sigma 500 and the Nikkor 500 extensively...and, once again, I'm just going to repeat my the three sentence summary found in the optical stabilization section of this review (above):
There was extreme similarity in the shutter speeds at which I could hand-hold the Sigma Sport 500mm f4 and Nikkor 500mm f4E VR at and still obtain both very sharp shots and slightly less sharp "keepers" when shooting bursts of 3 shots. When I shot several longer (10 frame) bursts of shots using the various stabilization settings on the two lenses I did find some differences between the effectiveness of the settings and the lenses. I obtained a slightly higher number of sharp shots and overall number of keepers with the VR settings on the Nikkor 500 compared to the OS settings available to a Sigma 500 user without access to a USB dock and Sigma Optimization Pro software (i.e., when using the default OS "view" settings on the Sigma lens), but this difference disappeared when I used the Sigma lens with one of its OS customization settings (OS Moderate View).
D. Autofocus Performance
So...what about AF performance? Can a third party lens maker FINALLY match one of the "big guns" in the Holy Grail of AF performance? The short answer is yes. The slightly longer answer was summarized in the AF Performance section above...here it is again:
The differences in AF performance of the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR are so subtle that they are unlikely to be noticed under normal day-to-day shooting; both lenses show a very snappy initial focus acquisition, both shift from closest focus to distant focus very fast, and both re-focus quickly and smoothly enough that they rarely miss focus on even fast moving subjects. Repeated trials of continuous high-frame rate shooting on a rapidly moving subject showed "keeper rates" of almost 90% for both lenses, but with the Sigma Sport having a slightly higher rate of sharp shots. Both lenses exhibited high focus accuracy on all 55 selectable focus points of a Nikon D5, but the Sigma Sport 500mm f4 showed a higher degree of focus accuracy on several of the outermost focus points on a Nikon D500.
The longest (but most thorough) answer? Scroll up to the section on Autofocus Performance!
E. Sigma Lens Customization and Optimization
I mentioned the ability of a user to modify (customize) the performance of the Sigma lens via using their USB dock and free Sigma Pro Optimization software several times above. But, because it was a "Sigma-only" feature - and because it is only available to users who purchase the optional USB dock - I didn't dedicate an entire section to it in this comparative review.
Here's a bare bones explanation of how Sigma's customization/optimization system for their lenses works: Those who purchase the optional USB dock and install Sigma's free Optimization Pro software can modify their 500mm f4 lens in several ways - they can upgrade its firmware themselves, they can input multiple AF fine-tuning values for different subject distances (for a SINGLE camera), they can set-up how much focus-ring rotation is needed to shift the lens to manual focus mode, and they can set up two different custom setting banks that can the user can toggle between using a switch on the lens. Within EACH custom setting you can tweak the AF Speed setting (Fast AF Priority vs. Standard AF vs. Smooth AF Priority), set the Focus Limiter Setting (i.e., limit the distant range the lens can focus to), and tweak the Optical Stabilization Setting (Dynamic View Mode vs. Standard vs. Moderate View Mode).
I am going to delve into many aspects and nuances of the customization capabilities of the Sigma "system" in a future blog entry, but for now there's a few things relevant to this review that need saying:
• I REALLY like the entire concept - all aspects of it make sense to me...from the user-updatable firmware capabilities to the tweaking of the customization settings. I don't think ALL aspects of it are "perfect" yet (e.g., owners of Sigma lenses who have multiple cameras may not like the fact that their AF tuning is tied to a single camera), but the concept makes sense to me.
• Users who want to buy a lens, take it out of the box, and "just use it" (and not spend a bunch of time monkeying around with a bunch of settings that they may struggle to tell the difference between) might hate it (or choose not to buy into the system...which could leave them with a lens that does NOT match the competition...see immediately below).
• I was surprised to find that tweaking of the customization settings can make a BIG difference to the performance of the lens. In fact, parity with the Nikkor 500 in AF and "hand-holdability" was only experienced after I tweaked AF speed and OS custom settings. In my view (= opinion!) anyone buying the Sigma 500 should buy the USB dock. Actually...in my view the USB Dock is so essential that it should be included with the lens (even if the lens has to go up by $50 or so in price).
• It's my opinion that IF the target market for this lens is primarily wildlife and sports photographers, then the default custom settings aren't the best. Because I KNOW I will get the question, I am finding that my preferred custom settings (my new "defaults") for the Sigma 500 are
• For AF Speed Setting: Fast Priority AF
• For OS Setting: Moderate View Mode
II. MY 500mm LENS DECISION
I had two decisions to make when I started this test. First, would my wildlife kit benefit from having a 500mm lens in it (given I own Nikon's 400mm f2.8E VR)? The answer to that question is yes.
The second question - which lens? I have decided to keep the Sigma Sport 500mm f4.
So...why do I need BOTH a 400 f2.8 AND a 500mm f4? The biggest reason is the fact that the lens is smaller (especially in front element diameter) and lighter for travel AND for carrying in the field in my backpack. In the course of my work I occasionally travel via commercial flights, float planes, and even helicopters. Invariably I am up against the limit in luggage size or weight allowances, so saving over a pound does make a difference. Of course, at times the extra reach of a 500mm (over a 400mm) without having to add in a teleconverter will prove useful, especially given that at most apertures I found both 500's to be slightly better optically than the 400mm f2.8E VR with the TC-14EIII (1.4x) attached to it.
So...why the Sigma over the Nikkor? First, I feel compelled to say that a purchaser wanting a high-quality super-telephoto lens can't go wrong with either of these two lenses. And, for me BOTH earned enough of that nebulous factor I call "gear confidence" (which boils down to NEVER feeling hesitant to use a particular piece of equipment) that I would be VERY happy with either of them. And, of course, in my mind they are in a dead heat in optical quality, AF performance and "hand-holdability" (which together end up making the two lenses absolutely equal in "usability"). But here are my main reasons for selecting the Sigma Sport 500mm:
A. Slight "Nebulous" Edge in Build Quality: OK...this reason IS subjective. And don't ask me to explain it further - but to me the Sigma just FEELS more robust and durable. Ironically it might be the increased lens weight relative to the Nikkor. Or, I might be biased a bit by the fact that I have found the Sigma Sport 150-600 zoom to be absolutely bomb-proof and NEVER fail in the toughest conditions imaginable. Maybe I'm being fooled by the "Soviet-era" inspired cosmetics on the Sigma. And, I really like that all rings on the lens are just so super smooth in movement (with the most significant example being on the action of the lens collar...just SO smooth on the Sigma).
B. A Few Minor Differences in Physical Features: Sigma has done a few little things better than Nikon, including offsetting the AF Activation buttons on the lens and adding in "detents" (that you can turn on or off) on the rotating lens collar. Just real nice LITTLE touches.
C. Lens Customization: I think this is a GREAT feature. It may be more important to me because I live in a remote location - when I hear that the Nikkor 200-500mm f5.6 zoom needs a firmware upgrade AND I have to send it to Nikon to get the upgrade, I'm looking at being without the lens for two weeks (or, more likely, just accepting that I won't bother and will have to learn to live without the benefits of the firmware update). With the Sigma 500 it's a 5 minute fix I can do myself. Sweet. And, as explained above, I have found the customization features useful as well. I prefer the idea of buying into an upgradable and evolving (and improving) product over buying what amounts to a static product.
D. Service and Corporate Responsiveness: OK...since I began this test I have come up with TWO issues on the Sigma lens that required contact with Sigma. The first pertained to the customization feature. Initially the Sigma Optimization Pro software simply didn't "see" the 500mm lens when it was attached to the USB dock. I immediately reported this problem to Sigma - and it was just a few days before Christmas. They instantly (as in same day) got back to me and said they would check it out and attempt to find a solution as fast as possible. Of course, getting ANY kind of special service between Christmas and New Years from ANYONE is normally next-to-impossible. Nonetheless, by January 3 they got back to me and had fixed the problem. Next time I connected the USB Dock and started the Sigma Optimization Pro everything worked perfectly. What? Good service from a Japanese camera and lens company? Mind-boggling.
The second issue pertained to the focus-shift issue (between frames in a high-speed burst on a static subject) I reported in my AF Performance section. When I contacted Sigma with this problem they quickly got back to me and let me know they were able to replicate the problem (based on the details I gave them) AND that they were working on the issue. At the time of this writing they haven't solved the problem yet, but I am confident they are trying. WOW. No denials. No brick walls. How refreshing!!
E. BUT, BUT, BUT...What About Lens Weight? Yep, I wish this lens was as light (or lighter) than the Nikkor 500mm f4E VR. But I'm not tiny and have lived through owning and hand-holding the VERY HEAVY G versions of the Nikkor 400mm f2.8 VR and 600mm f4 VR. I can live with the 324 gm (0.71 lb) penalty of the Sigma Sport 500mm f4.
And...that's a wrap. I'm now grabbing my D5, Sigma Sport 500mm f4, and going shooting. Have a good day!
Appendices
FULL STOP: The following appendices are included for those wishing more details on selected sections of the performance comparison of the Sigma and Nikon 500mm lenses. They are not at all essential reading - one can get a full understanding of how these lenses compare WITHOUT delving into this level of detail.
APPENDIX 1. Optical Performance - MORE DETAILS on Procedure and Results:
Boring Alert! This section may bore some readers to tears, so feel free to ignore it. There are a number of small gems contained within, but the primary conclusions have already been mentioned in the EXECUTIVE SUMMARY. If you're detailed oriented...well...read on!
A. Testing at 7 Meters (D5) and 10.5 Meters (D500)
I chose these distances because they represents the working distance I (and presumably many other nature photographers) use with small mammals such as squirrels and chipmunks and medium-to-large songbirds, such as American Robins, various jay species, et cetera. The target I chose was a stump located in my yard that has good wood-grain detail on it (making it easy to see differences in sharpness) as well as having exhibiting a mix of slightly out-of-focus (or OOF), more OOF, and completely OOF zones (thus providing an opportunity to assess the quality of the OOF zones for each lens at each aperture). Here's a full-frame shot of the subject (with resolution reduced to 2400 pixels in Photoshop CC 2017):
• The Subject Stump (with D5 & Sigma Sport 500mm lens @ f8): Download 2400-pixel image (JPEG: 1.2 MB)
Please note that in this portion of the test it was impossible to assess edge sharpness (edges on this test subject are far in the distance and thus completely OOF). I'm not bothered by this because when I am working at close distances to my subject I am thinking MORE in terms of the centre sharpness with the background (commonly on the edges) soft. Think "portraiture".
At this distance I tested the Sigma Sport 500mm against 3 other lenses - the Nikkor 500mm f4E VR, the Nikkor 400mm f2.8E VR, and the Sigma Sport 150-600mm zoom (@500mm). The prime lenses were tested using both Nikon D5 and D500 and both when the prime lenses were shot native and with their respective teleconverters in place. Images were captured at 1/3 stop increments from wide open (f4 for the two 500's, f2.8 for the 400mm prime, and f6.3 for the Sigma Sport zoom) through to f8, and then at 1-stop increments through to f11. Images were captured in a highly controlled fashion (more controlled than during most wildlife shooting sessions) as described in the Introduction section (above)...using Live View, mirror-up, firm tripod, electronic front shutter curtains, et cetera. Note that for each aperture for each lens/camera combination I captured TWO images separated by about 10 seconds (and re-focused the camera between shots). This last step was taken because regardless of how careful one is, at times focus can be "missed" be enough to influence the results (and during my image assessment phase I compared the two images shot at each aperture and selected the sharper of the two...though in most - but not all - cases they were equally sharp).
Note that the test images using the D5 were captured at 7 meters to the subject and those using the D500 were captured at 10.5 meters to the subject (which is 1.5x times larger - the same as the crop factor of the D500).
I assessed image quality via examining previews of the shots using both Lightroom CC and Capture One Pro 10. Images were viewed at 100% magnification (1:1) on a 30" Apple Cinema Display with a native resolution of 100 pixels per inch (ppi). I chose this monitor for all image comparisons because small differences in sharpness are often "masked" when images are viewed at 100% on some newer high resolution displays (e.g., almost ANY image looks sharp on my 218 ppi iMac 5K Retina Display). In simple terms, I was pixel-peeping!
RESULTS AND OBSERVATIONS (@ 7m)
As expected, the D5 and D500 camera bodies produced virtually identical results and trends. For simplicity's sake I am reporting and discussing only the results of the D5 and various lenses shot at 7 meters.
1. Sigma Sport 500mm f4 (native): This lens exhibited the "typical" high-end super-telephoto trend of being slightly less sharp when shot wide open (f4), but sharper as you stop down just a little. In this case at f4.5 the images were slightly (but noticeably) sharper and then slightly sharper again at f5. However, stopping down further produced almost no noticeable increase in sharpness. Interestingly, even on the images shot with the full-frame D5 in diffraction at small apertures (including f16) wasn't much of an issue - the f16 shots were slightly softer than the f11 (or f8) shots, but quite close to the f5.6 images in sharpness. Similarly, there seemed to be virtually no observable chromatic aberration issues (and there were white edges against darker backgrounds...where any color-fringing most commonly shows). OOF zones were smooth and "buttery" looking...and of the high quality you'd expect of a high-end super-telephoto.
2. Sigma Sport 500mm f4 plus Sigma TC-1401 (1.4x) Teleconverter (700mm focal length): Again, the same "stop down 2/3 of a stop to get sharpest results" trend was obvious, but in this case that meant 2/3 of a stop from f5.6 - which means you want to shoot this lens plus teleconverter at f7.1 or smaller apertures to get sharpest results. And, at all apertures there was a very slight softening of the image with the teleconverter on. So, if you stack up the slight image softness when shot at the largest apertures AND the slight image softness associated with the presence of the teleconverter itself, I am left feeling that I would only rarely shoot the Sigma 500mm f4 with the TC-1401 teleconverter at apertures larger than f7.1.
3. Nikkor 500mm f4E (native): IDENTICAL comments to that of the Sigma Sport 500mm shot native (re-read if necessary).
4. Nikkor 500mm f4E plus Nikon TC-14EIII (1.4x) Teleconverter (700mm focal length): Again, IDENTICAL comments to that of the Sigma Sport 500mm shot native (re-read if necessary).
5. Sigma Sport 500mm f4 VERSUS the Nikon 500mm f4E VR: These two lenses could easily exhibit the exact same trends in optical performance (shot native or with teleconverters) but still differ in absolute sharpness. But if they DO, I was unable to find ANY differences in sharpness (at any aperture) OR in the quality of the OOF zones. To be clear, the lenses performed virtually identically (optically) under the controlled conditions the images were captured under. Optically my two copies of the lenses (at this distance) were like clones.
6. NO Sigma-Nikon Differences at ALL? Well...I found one small one. When I captured these shots the sky was overcast, and very early on I noticed that the images shot with the Nikkor 500 were always slightly cooler than those shot with the Sigma 500 (and please note that I was using Auto WB on the D5 and D500). During subsequent formal and informal shooting sessions I observed the same trend. Note that these WB differences (Nikkor cooler; Sigma warmer) were very small and were not apparent when I shot the lenses in full sunlight (and, of course, if one is a raw shooter this capture difference in WB can easily be "adjusted away" during raw processing). I am not sure of the source of the WB difference, but my best guess is that it reflects different coatings of the lens elements used by Nikon and Sigma. The degree of difference can be seen in the sample image below.
What about focus breathing (where some lenses shorten in focal length when focused at close distances) - was there any Nikon-Sigma difference? Nothing significant (my images showed under a 1% difference in the number of pixels dedicated to the subject width, which could easily be explained by minute differences in positions of the lenses on the gimbal head). Please note that I am NOT saying these two 500's exhibit no focus breathing - simply that if you compare the size of the subject within the frame of an image captured at 7 meters there is no obvious DIFFERENCE in the focus breathing of the two lenses (they BOTH could be breathing, it just so happens that it's by the same amount).
7. Sample Images? Stump images are real boring, but here's a (shot at f6.3) that shows the LARGEST difference I could find in image quality. Best to view the image at 100% magnification on a standard resolution (i.e., non-Retina or non-HD) display to search for differences in sharpness):
• Sigma Sport 500mm vs. Nikkor 500mm f4E VR @ 7 Meters: Download Comparison Image (JPEG: 1.8 MB)
8. The TWO 500's vs. the Nikkor 400mm f2.8E VR? I've often wondered if there is really any net benefit of owning BOTH a 400mm f2.8E VR and a quality 500mm f4 lens, especially given how well the Nikon 400 pairs up with the TC-14EIII (550mm focal length) and the TC-20EIII (800mm focal length). So I decided to take the opportunity during this test to also collect sample images shot with the 400mm f2.8E VR and with the 400mm f2.8E VR PLUS the TC-14EIII teleconverter (500mm focal length) and then digitally resize BOTH sets of images to match a 500mm focal length. In the case of the images shot at 400mm I processed the raw images as normal (using Capture One Pro) and then upsized the images in Photoshop CC 2017 using the "Preserve Details" image sizing algorithm to match the subject dimensions (in pixels) of the images captured with the 500mm lenses. With the images shot at 550mm (i.e., with the TC-14EIII teleconverter) I processed the raw images as normal and then downsized the images in Photoshop CC 2017 using the simple "Bicubic" (not Bicubic Sharper, not Bicubic Smoother) algorithm to match the subject dimensions (in pixels) of the images captured with the 500mm lenses.
What did I find? First, that the images shot at 400mm and then UPSIZED (UPSAMPLED) in Photoshop didn't fully match the sharpness of the images of the Sigma 500 or the Nikon 500. And, they had increased (and, in my view, excessive) contrast. This was true at all overlapping apertures, with the images that were most comparable in overall quality (especially sharpness) being the ones where the 500mm lenses were at their weakest, i.e., shot wide open at f4. So...at this point in time a great 400mm lens and image upsizing doesn't quite give you 500mm lens quality (at least at close distances).
Second, that the images shot with the 400mm f2.8 lens PLUS the 1.4x TC and then downsized in Photoshop stacked up quite well against the images shot with both 500mm lenses. In fact, if you looked ONLY at image sharpness, the images shot at f4 with the 400 plus TC (then downsized) were slightly better than those shot with both the Nikkor and Sigma 500's. However, by f5 and beyond the images shot with the 500's were sharper than the 400 plus TC (downsized). And, at ALL apertures the OOF zones of both 500mm lenses were smoother (and, at least for me, more pleasing) than the OOF zones of the 400mm plus TC (I have noted this before in my teleconverter reviews, i.e., that sometimes the greatest image degradation with a TC isn't in the image sharpness, but rather the negative impact on the OOF zones...with OOF zones of images shot with TC's being more "jagged" and "nervous" than the OOF zones of primes shot native).
9. How Did The Sigma Sport 150-600 (@ 500mm) Stack Up? Not too bad, although at a subject-to-camera distance of 7 meters the Sigma Sport 150-600mm exhibited enough focus-breathing that it made an absolute sharpness comparison challenging (at 7 meters the width of the subject stump had 7% fewer pixels, making it more akin to a 470mm lens). That being said, it was still obvious that the images shot with the two 500's were considerably sharper up to and including f7.1. At f8 and beyond the images shot with the Sigma Sport 150-600 were quite close in sharpness to those shot with the two 500's. However (and possibly owing at least partially to the focus-breathing issue of the 150-600), the OOF zones of ALL the Sigma Sport 150-600 were very noticeably less smooth than those shot with either of the two 500's (and at this distance this difference in the bokeh was easily noticeable at all apertures).
B. Testing at 30 Meters.
I chose this distance as my "next" testing distance as it is quite representative of the distance I often work at with larger subjects, including many species of mammals (bears, wolves) and larger birds (such as owls and eagles), and often even including birds-in-flight. For me (and I suspect many wildlife photographers) the optical performance in the 20-50 meter range is extremely important. The target I chose at this distance WAS a Bald Eagle, albeit a life-sized one carved out of wood! As with the 7 meter distance, I chose an angle that included background objects at several different distances, which facilitated comparing the quality of partial and fully OOF zones.
Here are full-frame shots of the subject (with resolution reduced to 2400 pixels in Photoshop CC 2017) captured taken with the Sigma Sport 500 and the Nikkor 500mm f4E VR (both images shot at f7.1; ISO 400; 1/320s). It is acknowledged this is a butt-ugly scene with uninspiring light...but it was useful for testing purposes! ;-)
• Sigma Sport 500mm @ 30 meters (f7.1 on D5): Download 2400 pixel image (JPEG: 1.25 MB)
• Nikkor 500mm f4E VR @ 30 meters (f7.1 on D5): Download 2400 pixel image (JPEG: 1.25 MB)
As per the 7 meter test my concern at this distance was sharpness of the subject, not edge-to-edge sharpness (and again this scene did not lend itself to assessment of edge sharpness).
Image capture protocol as per at 7 meter distance (Live View, etc.). Image assessment and processing as per 7 meter distance.
Note that at this distance I tested the lenses ONLY on the Nikon D5 (and I have no reason to believe the results and/or trends would be any different on different Nikon bodies).
RESULTS AND OBSERVATIONS (@ 30m)
1. Same Old, Same Old! The overall trends observed at 7 meters were repeated at 30 meters. Essentially there was optical parity between the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR (both extremely sharp with very similar bokeh, and with the same trend with the softest images being at f4, but both lenses approaching maximum sharpness by f5). And, the images from both 500's were superior in sharpness than those captured with the 400mm f2.8E VR and then upsized to the same magnification as those shot with the 500's.
2. Any Change in Results at ALL? There were a few small differences in the results at 7 meters vs. 30 meters. First, while the images shot with the 400mm f2.8E VR PLUS the TC-14EIII (and then downsized to match the magnification of the 500mm lenses) were STILL better than the images shot with the 400mm f2.8E and then digitally upsized (to 500mm), there was a bigger quality gap between the images captured with both 500mm lenses and the downsampled 400mm f2.8E VR plus TC-14EIII images. In short, the images captured with both 500mm lenses were sharper at all apertures than those shot with the 400mm f2.8E VR plus the TC-14EIII and then downsampled to 500mm - including at f4.
Second, the Sigma Sport 150-600mm (shot @ 500mm) fared less well at this distance - the difference in sharpness between both 500mm lenses and the Sigma Sport 150-600mm (@500mm) was much more noticeable.
Focus breathing at 30 meters (with any of the lenses)? Pretty much a non-issue at this distance. The subject height (in pixels) when I compared the two 500's was virtually identical. And, the higher degree of focus breathing on the Sigma Sport 150-600mm disappeared. In fact (and quite surprisingly to me), the subject size (in pixels) on the Sigma Sport 150-600mm was slightly larger than with both 500's (not a lot...only about 1.5% larger). At this distance the Sigma Sport 150-600 (set and "clicked into place" @ 500mm) seemed to slightly lengthen! I found this result so surprising that I questioned its accuracy - so I reconstructed the setup again and took a few shots with each 500 and the Sigma Sport 150-600 (@ 500mm) and got the same result again. Compared to the two 500's the Sigma Sport 150-600mm lengthened slightly at 30 meters. Go figure!
C. Testing with Distant Scenes (1.95 km)
I then jumped up to shooting distant scenes - in this case a rocky ridge 1.95 km west of my home. I often shoot "animalscapes" where the subject animal is a small part of the overall scene, and these are often with quite distant subjects (often in the 500 meters to 2 km range). Because at THIS type of distance I am normally concerned with showing an animal in a complete scene, I normally want the entire scene quite sharp, including the edges. Consequently I chose to shoot this scene with each of the cameras I might select - the D5, D500 and the D800e (the latter of which is quite demanding and tends to show any lens flaws).
Here are full-frame shots of the subject (with resolution reduced to 2400 pixels in Photoshop CC 2017) captured taken with the Sigma Sport 500 and the Nikkor 500mm f4E VR (both images shot at f5.6; ISO 100; 1/640s).
• Sigma Sport 500mm @ 1.95 kilometers (f5.6 on D5): Download 2400 pixel image (JPEG: 1.8 MB)
• Nikkor 500mm f4E VR @ 1.95 kilometers (f5.6 on D5): Download 2400 pixel image (JPEG: 1.8 MB)
Image capture protocol as per at 7 meter distance (Live View, etc.). Image assessment and processing as per 7 meter distance.
RESULTS AND OBSERVATIONS (@ 1.95 km)
1. Groundhog Day - All Over Again! The overall trends observed at 7 meters and at 30 meters were repeated with the subject at 1.95 km. That means almost stunning optical parity between the Sigma Sport 500mm f4 and the Nikkor 500mm f4E VR (both extremely sharp, and with the same trend with the softest images being at f4, but both lenses approaching maximum sharpness by f5). And, the images from both 500's were superior in sharpness than those captured with the 400mm f2.8E VR and then upsized to the same magnification as those shot with the 500's. And, the images shot with the 400mm f2.8E VR plus the TC-14EIII (550mm) and then DOWNSAMPLED in Photoshop were still strong - not quite as sharp as the images captured with either 500mm, but darn close.
2. Any Change in Results at ALL? Only one worth mentioning. At this subject distance the Sigma Sport 150-600mm "bounced back" some - while the Sigma Sport 150-600mm shots (@ 500mm) were soft at f6.3 and f7.1, by f8 (and beyond) they were almost as sharp as the images shot with the two 500mm lenses. Based on my experience with many other "super telephoto" zooms over the years (many of which tend to "falter" at very long camera-to-subject distances) this is a strong result.
Any obvious focus breathing on the lenses at this distance? Nope...not an issue.
3. Edge Sharpness Differences? Here I visually evaluated the edge sharpness on images shot with BOTH 500's and the D5, D500, and D800e. It's my experience that if ANY of these three cameras are going to reveal differences in the optical quality of the Nikon 500mm and the Sigma 500mm it will be the D800e. And that camera showed the same thing as both the D5 and D500 did, i.e., that both 500mm primes showed excellent edge-to-edge sharpness. Note that while some other super-telephoto primes (such as the 400mm f2.8E VR which is incredibly sharp across the frame with distant subjects) show good edge-to-edge sharpness, not all do. For instance, my copy of the Nikkor 600mm f4G prime lens was fabulous at short and medium distances, but had very soft edges if you pointed it at distant subjects.
APPENDIX 2. Stabilization and "Hand-holdability" - MORE DETAILS: RESULTS
Here are the detailed results found during my testing on optical stabilization and "hand-holdability". Again...the most significant findings are described ABOVE in the section on Stabilization and "Hand-holdability" so there is little need for most viewers to wade through this information. But...it's here for those keen on detail!
A. 3 Frame Burst Shot Testing:
OK - here I looked for two things. First, how slow of a shutter speed could I shoot at and still get "Consistently Sharp Shots"? I defined Consistently Sharp Shots as two of three shots in the burst had been categorized as Sharp (as per the categories above). Second, what shutter speed could I go down to and still get ANY "Keepers" (so at least ONE of three shots in the burst either Sharp or only Slightly Soft). Here are my results:
1. Shutter Speeds Necessary For Consistently Sharp Shots:
NIKKOR 500mm f4E VR
• VR OFF: 1/500s
• VR SPORT: 1/160s
• VR NORMAL: 1/125s
Sigma Sport 500mm f4
• OS OFF: 1/400s
• OS1 Standard View: 1/125s
• OS1 Dynamic View: 1/200s
• OS1 Moderate View: 1/125s
2. Shutter Speeds Needed For AT LEAST One Keeper Per 3-shot Burst:
NIKKOR 500mm f4E VR
• VR OFF: 1/160s
• VR SPORT: 1/80s
• VR NORMAL: 1/100s
Sigma Sport 500mm f4
• OS OFF: 1/125s
• OS1 Standard View: 1/80s
• OS1 Dynamic View: 1/160s
• OS1 Moderate View: 1/80s
3. Stability of Image Through Viewfinder (BEFORE shooting):
What I am referring to here is simply how stable the image appears through your viewfinder BEFORE you actually shoot. This isn't necessarily related to how sharp the final image may end up, but can be important in image composition. It may also help you keep your focus point on the EXACT spot you want it. And, a high degree of stability in the image through the viewfinder certainly reminds you the system is on and working, and may influence your perception of how effective the optical stabilization of a lens is (correctly or otherwise!). Admittedly I have no objective measure of this - it is purely subjective but - at the same time - was extremely obvious.
With the VR/OS systems OFF I had a devil of a time with both lenses in keeping the image through the viewfinder stable and the focus point on a particular same spot on the subject. It was only slightly better with the Sigma lens when using OS1 Standard View (the lens's default setting) and a little better again using OS1 Moderate View. With OS1 Standard View mode some might think the OS system wasn't really operating (it was, it just didn't really LOOK like it was!). Most shooters should notice improvement of the stability of the iimage through the viewfinder when using OS1 Moderate View mode. The three remaining modes - Nikon VR SPORT, Nikon VR NORMAL, and Sigma OS1 Dynamic View - were characterized by having high (and very similar) image stability through the viewfinder - you definitely knew the stabilization mode was working! As you'll see below (and partly by looking at the results above) the stability of the image as seen through the viewfinder is only poorly correlated with how effective the optical stabilization system really is.
4. Stability of Images BETWEEN Frames within a Burst:
Even with short 3-frame bursts it was obvious ONE mode (Nikon VR NORMAL) was different than all the others, including when the stabilization systems were off. Simply put, the image jumped all over the place between frames when using Nikon's VR NORMAL, and much more so than even when the VR (or OS systems) were turned off. This was obvious both while looking through the viewfinder and when scrutinizing images after the fact. If one was shooting only single shots this major between-frame bouncing would be irrelevant (it actually wouldn't exist), but when shooting bursts it's a MAJOR issue for me.
Please note that this between-frame "herky-jerkiness" is most obvious when using a Nikon D5. This is because it has the shortest mirror black-out time and its new mirror-driving mechanism is incredibly smooth - so when one is using VR SPORT mode on a lens with a D5 body the image stability is amazing. Even with the D500 (which also has a new mirror-driving mechanism, but it's less effective than that of the D5) the difference in between-frame shifting of the image between the two VR modes of the Nikon 500 is slightly less noticeable.
B. 10 Frame Burst Shot Testing:
This is one of those "Where do I begin?" sections! First off, there was extreme consistency between the 4 repetitions of this test (which helped convince me the trends were real). Consequently, I lumped the test results together before calculating the percentages in each category you'll see below. This part of my testing produced just under 2800 images to compare and scrutinize and I have a huge number of results. For the sake of simplicity I will present and discuss only the 5 areas that I think most shooters would find most relevant.
1. Lowest Shutter Speed Needed To Obtain 50% or More SHARP Shots Per Burst
NIKKOR 500mm f4E VR
• VR OFF: 1/500s
• VR SPORT: 1/125s
• VR NORMAL: 1/125s
Sigma Sport 500mm f4
• OS OFF: 1/500s
• OS1 Standard View: 1/160s
• OS1 Dynamic View: 1/200s
• OS1 Moderate View: 1/125s
2. Lowest Shutter Speed Needed To Obtain a 100% Rate of KEEPERS (SHARP and SLIGHTLY SOFT shots) Per Burst
NIKKOR 500mm f4E VR
• VR OFF: None of the tested shutter speeds - so a shutter speed FASTER than 1/500s
• VR SPORT: 1/125s
• VR NORMAL: 1/100s
Sigma Sport 500mm f4
• OS OFF: 1/500s
• OS1 Standard View: 1/200s
• OS1 Dynamic View: 1/250s
• OS1 Moderate View: 1/125s
3. Lowest Shutter Speed Needed To Obtain a 50% Rate of KEEPERS (SHARP and SLIGHTLY SOFT shots) Per Burst
NIKKOR 500mm f4E VR
• VR OFF: 1/400s
• VR SPORT: 1/60s
• VR NORMAL: 1/40s
Sigma Sport 500mm f4
• OS OFF: 1/400s
• OS1 Standard View: 1/160s
• OS1 Dynamic View: 1/125s
• OS1 Moderate View: 1/60s
4. Stability of Image Through Viewfinder (BEFORE shooting):
Exactly as reported for the testing with 3-frame bursts: Both VR modes on the Nikon 500mm f4E VR stabilizes the image you see through the viewfinder significantly, whereas with the Sigma Sport 500 the ONLY setting that produces highly stable images as seen through the viewfinder is OS1 Dynamic View.
5. Stability of Images BETWEEN Frames within a Burst:
Again, exactly as reported for the testing with 3-frame bursts (but even MORE pronounced): All the OS1 view modes of the stabilization system of the Sigma lens are very smooth between frames within a high-speed burst, and the VR SPORT mode is absolutely silky smooth between frames in a burst (as in "rock solid" in the viewfinder with the D5). In contrast, the VR NORMAL mode exhibits very noticeable between-frame jumping of images in a high speed burst (to call it very "herky-jerky is an understatement).
APPENDIX 3. Autofocus Performance - MORE DETAILS: WHAT I DID and DETAILED RESULTS
This section gets into some of the nitty-gritty details of what I did and what I found in assessing AF performance. It does contain some observations that go beyond what is described above in the Executive Summary...but those with little time to spare can probably quit reading now! ;-)
First, a few technical notes that apply to virtually all of my AF testing...
• Prior to doing any AF testing on these two lens I fine-tuned their AF systems as described above in the section on AF Tuning.
• The majority of the AF performance assessments were made with using the lenses paired with a Nikon D5 camera. Selected tests were also performed using a Nikon D500 camera.
• Shortly after acquiring my Sigma Sport 500mm f4 lens I used the USB Dock and Sigma Optimization Pro software to customize the AF Speed Setting for "Fast AF Priority" (so all AF testing reported here was performed using Fast AF Priority). This setting is designed to "Prioritize autofocus speed to reach the focus point as quickly as possible" (this is the verbatim description in the dialog box of the Sigma Optimization Pro software). The opposite "extreme" in the AF Speed Setting is "Smooth Priority AF" which offers "...a slightly slower but very smooth autofocus, ideal for use with video." And, the default setting for the lens (i.e., the setting you would have to live with if you did not use the USB dock and Sigma software to customize the lens) is Standard AF which is functionally a mid-point between the other two settings and "...achieves both smooth operation and fast AF speed." As an editorial comment...this is another case where I think MOST users of this lens would want it focusing as fast as possible, and thus the default setting for the lens should be Fast AF Priority (rather than Standard AF).
A. Comparative AF Performance Observations When "Just Shooting"
1. What I Did:
I have been shooting with both of these lenses now for almost 3 months and have been watching for noticeable differences in AF performance in the field (and while scrutinizing the resultant images). The subjects shot during this time have included an array of stationary subjects (everything from distant scenes to people to my dogs and to bighorn sheep and other species of wildlife) as well as moving subjects. I've also spent time quickly swapping between lenses on a tripod and seeing if I can get at least a subjective answer to the following types of questions:
• Is there any noticeable difference in AF speed when shifting from a close to distant subject?
• Is there any noticeable difference in initial focus acquisition over different distances?
• Does one lens hunt for focus more than the other (including when focusing on low-contrast subjects against non-contrasting backgrounds)
2. What I Found:
This type of comparison (I won't even call it "testing") generally reveals only very large and obvious differences in the AF performance of two lenses - so "macro" differences only. But occasionally it does point out areas worth investigating further (which is EXACTLY what it did in my case...see the "Noticeable Idiosyncrasy" below).
The most germane finding is that I could detect virtually no difference in almost all aspects of AF performance when "just shooting" in the field - both lenses focus VERY fast on a subject, neither hunts for focus until one is trying to shoot in near dark conditions (of course, this is partly a function of the camera body the lens is paired with, but the key thing is that I could find no BETWEEN-LENS differences), both lenses can re-focus very fast when moving from closest focus point to a distant focus point, etc.
BUT...One Noticeable Idiosyncrasy! Part way through the test period I had occasion to shoot some close-up shots of squirrels with both lenses mounted on a D500 (both with and without their respective teleconverters). Think "full-frame shots" and think SHALLOW depth-of-fields (or DoF's). During this time I was shooting in short, high-speed (10 fps on D500; 12 fps on D5) bursts. And, when scrutinizing the shots, I noticed that with the shots taken with the Sigma 500 there was some inconsistency in focus between images captured within a single high-speed burst (when the subject is absolutely static and the camera is not moving). That lead me to shoot several sets of test shots to see what I could learn about this "focus shifting" between frames in a high-speed burst. Here's what I found out:
• The focus shift is normally only noticeable on static subjects
• The focus shift seems to only occur on high frame-rate bursts (greater than 5 fps or so)
• The focus shift is subtle and thus most noticeable on shots with very thin DoF's (e.g., close subjects)
• The focus shift pattern is often as follows: first shot sharp; second shot soft (under the focus point); third shot sharp, etc.
• The focus shift is independent of the OS and AF settings (both those that are adjustable on the lens OR through customization) - it occurs in all AF or OS modes.
• The focus shift occurred on both D5 and D500 bodies.
• The focus shift did NOT occur when I was using the Nikkor 500mm f4E VR
Example shots? Yep. Compare these two shots out of a 10 fps burst of a squirrel that had JUST paused. These are the FIRST and SECOND shots in the burst. Tech specs are on the shots, but for reference both shots with Nikon D500 and Sigma Sport 500mm f4 with TC-1401 teleconverter in place (so an EFL of 1050mm). While I was stopped down to f9 in these shots, the DoF is still VERY thin. In both shots the focus point was positioned half-way between the eye and the nose of the squirrel.
• Sample 1 (Sharp in facial region): Download 2400 pixel image (JPEG: 1.3 MB)
• Sample 2 (Softer in facial region): Download 2400 pixel image (JPEG: 1.3 MB)
After discovering this idiosyncrasy I contacted Sigma to report it. As I have found before, the Canadian distributor was very responsive and forwarded the information on to Sigma in Japan. They quickly got back to me with more questions about the conditions under which I noticed the focus shift. Shortly thereafter (days later) they got back to me again to report that they were able to replicate the issue (and that they had found it on some Nikon lenses as well, but did not indicate which lenses). While Sigma did not have an immediate "fix" for the issue, they assured me they were working on it.
How serious is this issue? For ME, not too serious at all. In most of my "normal" daily shooting situations it simply doesn't occur (or isn't noticeable). In situations where it has occurred (so far shooting small stationary mammals that are near the close-focus point of the lens) I have come away with the shots that I wanted (i.e., no critical shots were lost to the problem). I habitually use short bursts whenever I'm photographing wildlife, and I will try to remember to perhaps extend those bursts a LITTLE more if I'm in really tight (and/or just have thin DoF's) with my subjects. And, based on past experience, I am realistically (I think) and reasonably confident that Sigma will come up with a fix for the issue. And if they do, any owner of the Sigma 500 will be able to use their USB dock to upload the fix! ;-)
B. AF Testing on a Rapidly Moving Subject
Like so many wildlife photographers I love shooting moving wildlife. You know...birds in flight (BIFs) and running mammals. And, for me (and probably most wildlife shooters), if a super-telephoto can't maintain focus on a rapidly moving subject, I don't want it. Period.
So...in deciding between which of these two super-telephotos I wanted to keep I had to know if they could "make the grade" in keeping moving subjects in focus. Or, in other words, they had to excel in three distinct aspects of AF performance - autofocus speed, predictive AF capabilities, AND focus-tracking. Predictive autofocus capabilities (which most modern DSLR's have whenever one is in continuous servo mode) are important in keeping a subject that is moving away or towards the camera, even if it stays in position under a single AF focus point. Focus tracking is important if the subject is moving erratically and it's likely that you CAN'T keep the subject under a single focus point (or zone). In these cases the camera/lens combination has to be able to "hand" the focus off when it moves from one point to another (without losing focus on the subject in the process). And, of course, focusing speed is critical simply to "keep up" with the movement of the subject (especially if the subject is moving towards or away from the photographer).
I tested these three AF characteristics collectively by shooting continuous high-speed bursts of a "sort of" trained dog who very much likes to run directly at me at breakneck speed, followed by examining the number and percentage of sharp shots, keepers, and out-of-focus shots (all defined below) in each sequence of shots. I very much like this test because it places high demand on the AF system and it can be repeated time and time again. When a dog is running directly at you it is (of course) forcing the lens to continuously refocus as the distance to the subject decreases. And, because the dog's head is continuously bobbing up and down, it is virtually impossible to keep the dog's head under a single focus bracket and thus requires that the AF system "passes" the focus between focus points (i.e., focus tracking). I have been performing this test on various lenses for years and I have discovered that this test is much more demanding on the AF system than MOST BIF shots (unless, of course, one is trying to photograph full-frame shots of swallows hawking insects while on the wing).
1. What I Did:
Here are the technical details of the testing procedure:
IMAGE CAPTURE:
• I used a hand-held Nikon D5 body at its highest frame rate (12 fps)
• I independently tested (did separate trials) using 72-point dynamic area AF mode and 9-point dynamic area AF mode. I ran four separate trials with EACH of the two lenses for each area mode. So 4 trials with the Nikkor 500mm f4E using 72-point dynamic area mode, then 4 trials with the 500mm f4E using 9-point dynamic area mode and then the same all over again but using the Sigma Sport 500mm f4 lens. Given the distance covered by the trial (approximately 100 meters) each trial produced roughly 110 shots (I began shooting just before calling Poncho the Portie and ended them when he was more than filling the entire frame). In the end I ended up with a little over 1650 images to assess.
• All images were captured at f5.6 and 1/2000s (to keep the DoF fairly narrow AND to ensure the action was frozen to limit image softness to focus misses - rather than to motion blur). Auto ISO was used and the ISO varied slightly over the shots (both within and between trials)
• All images were captured as 14-bit compressed raw (.nef) files
IMAGE/SHARPNESS ASSESSMENT:
• All images were painstakingly (trust me, this got VERY boring!!) assessed for sharpness at 100% magnification in Lightroom and on a 30" Apple Cinema Display with a native resolution of 100 pixels per inch (ppi)
• Each image was placed into one of three categories - Sharp, Slightly Soft, and Unacceptable
• Sharp images were those where the leading edge of the subject (Poncho's nose) was absolutely sharp with all "nose wrinkles" visible
• Slightly Soft images show only VERY slight softening of the image and that softening is such that it can be overcome with careful sharpening (to be made indistinguishable from the Sharp images at full resolution)
• Unacceptable images are "all the rest" (which corresponds to the ones you'd trash during normal image culling)
• Keepers? This is all those that are NOT UNACCEPTABLE (Sharp Images + Slightly Soft Images = Keepers).
For reference, here are two shots (one shot with the Sigma Sport 500, one with the Nikkor 500) that are KEEPERS (and both were initially categorized as "Sharp"). In these shots you will see what I mean by "nose wrinkles":
• Poncho with Sigma Sport 500: Download 2400 pixel image (JPEG: 1.0 MB)
• Poncho with Nikkor 500: Download 2400 pixel image (JPEG: 1.3 MB)
2. What I Found (RESULTS):
An accurate overall description is simple: Both lenses performed superbly and produced extremely high keeper rates and very good rates of sharp images. And, the Sigma did just a little better than the Nikkor.
The numbers? Here you go. Note that because there was no significant difference between any of the 4 trials for a given set of conditions (i.e., for the same lens and same focus area mode) I lumped the trials together to increase the sample sizes):
i. For 72-point Dynamic Area AF Mode:
• Sigma Sport 500 (N=426): Sharp = 215 shots (50.5%); Slightly Soft = 169 (39.7%); Unacceptable = 42 (9.8%). So...Keepers = 384 (90.2%)
• Nikkor 500 (N=433): Sharp = 181 shots (41.8%); Slightly Soft = 196 (45.3%); Unacceptable = 56 (12.9%). So...Keepers = 377 (87.1%)
ii. For 9-point Dynamic Area AF Mode:
• Sigma Sport 500 (N=408): Sharp = 201 shots (49.3%); Slightly Soft = 161 (39.4%); Unacceptable = 46 (11.3%). So...Keepers = 362 (88.7%)
• Nikkor 500 (N=417): Sharp = 150 shots (36.0%); Slightly Soft = 204 (48.9%); Unacceptable = 63 (15.1%). So...Keepers = 354 (84.9%)
I found these results extremely interesting. While the Sigma Sport did slightly better than the Nikkor in holding focus on this rapidly moving subject (seems odd to call Poncho a "rapidly moving subject"!), to me the take home lesson is still one more of AF parity than AF difference. Nikkor super-telephoto lenses are known to have extremely effective AF systems, and any test results that show that a 3rd party lens has matched a Nikkor in real-world AF performance (and in this case slightly exceeded the Nikkor) is quite the accomplishment. Simply put, I did NOT expect that the Sigma 500 was going to be this good in keeping fast-moving subjects in focus.
C. Focus Accuracy Across the Array of Focus Points
Some lenses (including some 3rd party lenses) struggle to attain focus on some of the more extreme (non-central) focus points of modern DSLR's. This is especially true with some of the latest DSLR's, such as the D500 where the outermost focus points almost touch the edge of the viewfinder. In this test I evaluated the focus accuracy of all 55 selectable focus points of the D5 and the D500.
1. What I Did:
In this test I simply shot two images of a target (one using Live View [or contrast detect] AF and the second using viewfinder-based [phase detect] AF) and compared the sharpness of the two images. I then repeated the test for all 55 selectable focus points. In this test the shots captured using Live View serve as a control to compare against the shots captured using viewfinder-based AF.
Here are the key technical details of the testing procedure:
• All images shot from a Jobu Algonquin tripod and Jobu Heavy Duty MkIV gimbal tripod head. Shutter release on all Live View shots via a MC-20 cable release.
• All images shot with VR/OS systems OFF
• All images captured at 1/500s and f5 (with Auto ISO on and ISO floating between 100 and 280)
• Each test (for both cameras and both lenses and all 55 selectable focus points) were repeated twice
• The central "zero" point on a LensAlign long ruler (oriented at 45º to the plane of the image sensor) was used as the target for each focus point. At 7m the "red square" (see images below) surrounding the zero point corresponded closely to the size of the focus point on the D5 (and was slightly smaller than the focus points on the D500).
• All images were assessed for sharpness at 100% magnification in Lightroom and on a 30" Apple Cinema Display with a native resolution of 100 pixels per inch (ppi)
These two uncropped (but resolution-reduced) images should make the setup and the procedure used in this test clear:
• Sample Test Shot 1 (focused on LEFT-most focus point on centre row of D500): Download Image
• Sample Test Shot 2 (focused on RIGHT-most focus point on centre row of D500): Download Image
2. What I Found (RESULTS)
Like with the AF test of moving subjects, an overall description of the results is quite simple: Focus accuracy of the two lenses was very similar, with a slight advantage to the Sigma lens only when the test was done using a Nikon D500.
Here's some additional detail...
AF Performance Using Live View: As expected, ALL images captured with Live View (with both lenses, both cameras) were tack sharp.
AF Performance Using Phase Detect AF on Nikon D5: Focus accuracy was perfect (on ALL 55 selectable focus points) when using phase detect AF (viewfinder-based) AF with BOTH 500mm lens when mounted to a Nikon D5. This means that for every selectable focus point for BOTH the Nikkor 500 and the Sigma 500 the image was just as sharp when using the viewfinder-based focusing as it was when the same place on the subject was focused on with Live View focusing.
AF Performance Using Phase Detect AF on Nikon D500: When the Sigma 500 was mounted on a Nikon D500 it showed highly accurate (as in "perfect") focus on all 55 selectable focus points when using phase detect (viewfinder-based) AF. However, when the Nikkor 500 was mounted on a D500 it focused accurately on 51 of 55 focus points. The four points the Nikkor 500/D500 combination misfocused on were the FOUR extreme corners (see this graphic for details).
So...how much did the Nikkor 500/D500 combination misfocus on those 4 corner points? This image (shot with target square under upper right corner focus point) is representative of how out-of-focus all four misfocuses. It's best to view this image at 100% magnification to evaluate the sharpness difference between the control (Live View) and test (viewfinder-based) images:
• Sample - Nikkor 500mm/Nikon D500 Misfocus: (focused on LEFT-most focus point on centre row of D500): Download Image
In a sense I found this result even MORE surprising than the results from the testing of the lenses with a moving subject. My preconceived expectation was that if there was any difference at all between the lenses in focus accuracy between the two lenses the advantage would have gone to the Nikkor lens. I was wrong.
It's tempting to jump on results like these and be left with this simple thought: "Hah...the Sigma 500 BEAT the Nikkor 500!" But the reality is that there was NO difference in focus accuracy between the lenses when they were mounted on a D5. And, there was NO difference in focus accuracy in 51 of 55 focus points when the two lenses were mounted on a D500. Those 4 four points where the Nikkor 500/D500 combination missed focus are in some pretty extreme positions, and I suspect if I did not do this kind of almost anal testing I would NEVER have noticed the focus accuracy difference in the field (be honest - how often do you use those 4 extreme corner focus points when focusing a super-telephoto lens on a subject?). At longer distances to the subject (and smaller apertures) odds are one would NEVER notice that those 4 corner points don't focus quite as accurately as the remaining 51 on a D500!