An alternative to the Nikkor...
Nikkor 50mm on left, Sigma 50mm on right.
elements in 6 groups.
AF/Manual switch, HSM lens focus motor, internal
focus, 77mm filter size. Comes with petal hood, 9-blade aperture. Focuses to
About 2.7" (68.2mm) long, 17.8 ounces (505g).
50mm Sigma mounted on Nikon D60 DSLR
Before Nikon got around to updating their 50mm f/1.4, Sigma managed to sneak out their own version, the Sigma 50mm f/1.4 EX DG HSM. For a short time, it was the only fast fifty with an in-lens focus motor, though now it shares the table with the Nikkor 50mm f/1.4G AF-S.
Let me explain why a fast 50mm prime is still important in these days of zooms: for DX cameras, somewhere in the 50 to 60mm range is a perfect portrait focal length, and you generally want a fast lens to help provide control over depth of field (often you want to isolate a person you're taking a picture of from the background). An f/2.8 lens, or worse, a f/3.5-5.6 zoom, doesn't give as much flexibility at subject isolation. The older 50mm lenses use screw-drive autofocus systems that don't work on Nikon's lower-end models (D40, D40x, D60, and D5000). The Sigma 50mm HSM was the first to appear that provides autofocus with those consumer DSLR models, the Nikkor 50mm f/1.4G the second.
Unlike the Nikkor 50mm f/1.4G AF-S, the Sigma f/1.4 EX DG HSM packs some heft. Almost twice as much, as a matter of fact. At almost 3" long and with a 77mm filter ring at the front, this lens is dramatically bigger than the Nikkor. It's also heavier, weighing in at over a pound (about 7 ounces heavier than the Nikkor). Still, it's not a large lens, just larger than the most direct competitor. On a D60 or D5000 body the 50mm f/1.4 EX DG HSM adds substantial heft to the camera, though not enough to make for an unbalanced load. Personally, the Sigma feels "more right" on the front of a D60 than the Nikkor, about what you'd expect a portrait lens to be like. On a big D3-type of body, the Sigma balances nicely, but does make the body prone to tipping if you place it plate down on a flat surface.
The 50mm focal length provides 40° of horizontal angle of view on FX, 27° on DX. That 27° is pretty close to the 24° of the classic 85mm f/1.4D used on a film (or FX) body for portraits, thus my point about this being a good portrait length for DX. Just don't get confused and think that 50mm is "normal" on a DX body. It's not. It's a mild telephoto lens on DX.
Of course, FX body users will find that 50mm is "normal." The D700 and D3 have such good higher ISO abilities that I'm not sure you need an f/1.4 50mm any more for low light shooting, but dedicated night shooters will probably be happy with the fast aperture.
I mentioned depth of field isolation, so let's put that into perspective. On a DX camera taking a portrait at 10 feet you have about a half foot of depth of field at f/1.4, slightly over a foot at f/2.8. At 20 feet it's about two feet versus four feet. Put another way, you can cut depth of field in about half with the extra maximum aperture range at normal shooting distances, should you so desire. Personally, I find that level of extreme isolation a bit much, but there are most certainly photographers that value it.
The focus ring is moderate in size and easy enough to find at the front of the lens. Near to far focus takes just a quarter turn (and in the opposite direction of Nikkors). The lens has a distance scale with DOF markings for f/8 and f/16 for FX bodies (and at a looser standard than I'd use). On the left side (as you hold the camera) is one switch the controls focus (AF for autofocus, M for manual focus only). The lens does not have VR or an aperture ring (this last limits the number of older film SLRs that the lens can be used on).
The supplied hood is a petal type and can reverse mount on the lens. The pinch-front lens cap can easily be mounted on and taken off the lens with the hood on. The lens uses 77mm filters, a very large size for a normal prime.
The lens is made in Japan.
With a lens this simple, handling isn't exactly a deep subject.
The focus ring has a better, smoother feel than Nikon's, but has less precision with its quarter-turn range. Also, while you can kind of override autofocus manually, the lens will fight you on this much more than AF-S lenses do.
Weight can be a factor with this lens on small, consumer DSLRs, as 17 ounces is close to what the camera itself may weigh.
I don't like the Sigma rear lens caps: they are significantly more difficult to mount on the lens when you're in a rush or juggling equipment than are the Nikon caps.
You won't like the 77mm filter size on a "normal" (FX) or "portrait" (DX) prime lens, as this makes for expensive filters. If you're a filter freak, the Nikkor 50mm f/1.4G AF-S may be a better choice on just this point alone.
For a third-party lens, this Sigma falls into a middle price range, so you're probably expecting good performance.
Autofocus: the HSM focus motor in this lens does a better job than Nikon's AF-S motor in their 50mm. That's probably partly due to the shorter rotation on the focus ring. Whatever the case, the Sigma snaps a little faster to the focus point, and doesn't have the leisurely min to max (or max to min) sequence that the Nikkor does. If focus speed were your primary consideration, the Sigma easily wins the competition between the two lenses.
Sharpness: Like the new Nikkor it competes against, the Sigma 50mm f/1.4 EX DG HSM doesn't knock the ball out of the park. Indeed, looking at test results and images gave me a bit of deja vu. On DX cameras the corners were visibly softer at f/1.4 and the center wasn't pushing what the D300 could resolve. I had to stop down to f/5.6 to get the best possible results from both center and corner. On one hand, the Sigma seems to get to better center results faster (f/2 or f/2.8 depending upon where you set the bar for "excellent") than the Nikkor. On the other hand, the corners never quite come up to excellent levels, always being significantly softer than the center. On FX bodies the problems with the corners are just amplified. I'd say corners are decidedly poor at f/1.4 on the D3x, but then, what exactly are we expecting here? A modest priced lens with these specs isn't going to have exceptional corner performance wide open. Again, you'll need to stop down three or four stops to get to best performance on FX. Even then, performance never really reaches "excellent" in the corners on my D3x. If you're looking for best possible 50mm performance for a D3x, I'd tend to suggest the Zeiss manual focus lens. Don't get me wrong, performance is good, just not knock-your-socks-off great on a D3x. But again, what are you expecting putting a <US$500 lens on a US$8000 body? As my ratings try to suggest: you get what you pay for.
Here's what I got directly out of my D3x (FX body)
mostly set to defaults (test target is in center here, then moved to extreme corners for the corner tests).
This set of samples is exactly out of the camera with no manipulation other than to crop to the center of my test chart:
Compared to the Nikkor 50mm f/1.4G AF-S the Sigma is less sharp wide open in the center, about as sharp at f/5.6. But the Sigma ramps faster from the poor f/1.4 results to the good f/5.6 results. For example, f/2 is more usable on the Sigma. So it's a tradeoff: if you really need the f/1.4, the Nikkor is the better choice in the center. Beyond that, the Sigma has a slight advantage until about f/5.6, where both lenses are about as good.
In the corners, the Sigma never matches the Nikkor, always trailing a bit behind it. The Nikkor manages to get good results at f/5.6, but the Sigma still has low contrast and a bit more blur.
Also, I should note that for the samples I've just shown, I've placed the camera far enough from the test chart so that we barely get into chart patterns before we hit the Nyquist frequency (about "3" on most of the patterns). These samples show the lens at a distance I think it would be commonly used at rather than the closer distances normally used in resolution testing.
Light falloff: on the DX bodies vignetting is barely present wide open at about two-thirds of a stop. On FX bodies the vignetting is more noticeable and possibly enough that you might want to correct it, but in both cases it is negligible by f/2.8. Personally, only f/1.4 on a FX body is enough that I'd consider turning on Vignette Control or correcting after-the-fact. Light falloff is one area where the Sigma shows an easily demonstrable difference from the Nikkor 50mm 1.4G AF-S, especially on FX bodies.
from camera (FX)
Chromatic aberration: This is one place where I thought I was seeing deja vu: lateral chromatic aberration (side to side) is negligible wide open, with a tendency to slightly increase in visibility as you stop down, but it's never at levels I really worry about. More interesting is longitudinal chormatic aberration (near/far, sometimes called axial CA). The 50mm f/1.4 EX DG HSM most certainly exhibits such aberrations, especially near maximum aperture. Objects closer than focus get a magenta cast, objects further than focus get a green cast. This is not unusual for a fast prime, and the results are hauntingly similar to the Nikkor. Not a terrible flaw, but one to be aware of.
Flare: The big, slightly curved front element without much inset means you should use the hood, always. The hood does a reasonable job of keeping light from the front element, but I was able to get contrast reduction and a bit of flare on this lens that I didn't see in the same conditions with the Nikkor. Still, not really a problem you should worry about, just be aware that you've got to be a little more careful than you usually would with a 50mm to keep that front element from catching stray light.
Distortion: modest barrel distortion (<1%). Like the Nikkor, a bit high and more visible than you'd expect for a normal prime, but not enough to really worry about.
Bokeh: More deja vu. For the most part the out-of-focus highlights are basically circular as you'd want them to be (though they tended to get slightly elliptical on my sample in the mid-apertures). But that longitudinal chromatic aberration once again keeps me from fully praising the bokeh. Unlike the Nikon, which had pronounced colors at the edge of neutral highlights, the Sigma's are a bit more subdued. But unfortunately, they tend to produce a fairly clear ring at the edge of the highlight at maximum aperture. Stopped down a stop or two, the "extra detail" in the bokeh settles down a bit to the point where I'd say it becomes pleasant.
Compared to the Nikkor
The Sigma is bigger and heavier physically. Optically, the two lenses are close cousins in terms of performance. If central performance and bokeh (e.g. portraits) is most important to you, I'd slightly favor the Sigma. If overall performance is most important to you, I'd slightly favor the Nikkor. Autofocus performance is slightly faster on the Sigma at the expense of manual focus precision.
- No VR. This is an iffy call, but I think a lens that is going to be used a mild telephoto portrait lens by many probably ought to have had VR, because a lot of those portraits are going to be handheld.
- Longitudinal abberations. Colors aren't focusing to the same plane at maximum aperture, and is difficult to remove.
- Filter size. 77mm is a big, expensive filter size.
- Optically nice. Like the Nikkor f/1.4G AF-S, lots of good in the optics, with a couple of things to watch for. Probably slightly better for portraits than the Nikkor, slightly worse for edge-to-edge general use.
- Decent focus performance. Sigma's HSM outraces Nikon's AF-S in the 50mm dash.