Tiderace: i am completely dismayed that DP is owned by Amazon. I thought you folks were objective.
We're both owned by Amazon, *and* objective. The two are not in fact mutually exclusive.
Frank Petronio: Really how many times is the reviewer going to write "really"?
I'd blame your editor, really.
Paul Guba: Well if do consider buying one it will have nothing to do what is reviewed here. When your links are called campaign then you moved beyond editorial content into the realm of advertising and selling. I can't even read the review because I feel its an advitorial.
Oh, I now realise you're confused by the use of the word 'campaign' in the 'utm_campaign' section of the URL. This is simply Google Analytics code that allows us to track better how users progress through our site. It has nothing to do with advertising or product promotions at all.
(For example, you might see something like 'utm_ campaign=internal-link&utm_ source=mainmenu&utm_ medium=text&ref=mainmenu'. This simply tells us that the user arrived at the page from an internal link, using the main site menu. It's nothing more than that.)
Thanks for randomly providing a definition of the word 'advertorial', but it's completely irrelevant here - this review is not paid for by any external party.
That link isn't terribly relevant here, as it predates the Sigma 50mm F1.4 Art.
It isn't an advertorial, regardless of what you feel.
yoms: Hi!Can someone explain me something about the 3 shots of the distant clock to point out focus inaccuracy?
As the clock is really distant, I guess focus is set on infinity, right ? Moreover, focusing and exposure is always measured the aperture fully open, that is 1.4. So why is there misfocusing? Infinity is infinity, no ? There's no infinity + a little front focus or back focus.
Thanks for explaining.
So there are two points here. Firstly, it really doesn't take very much movement of the focus ring to blur the image by the amount seen in those shots. If you look at the [distance scale](http://www.dpreview.com/lensreviews/sigma-50mm-f1-4-dg-hsm/images/scale.jpg), there's a very small angle between infinity and the first marked distance, which is just 3m.
Secondly, depth of field calculations use a concept of 'acceptable sharpness' that's defined by the diameter of the 'circle of confusion'. For APS-C this is normally about 0.02mm, which is a blur circle covering 4 pixels. By these standards all three of those images would be considered to be 'in focus'. This reflects the fact that DOF calculations aren't about looking at 100% crops, but instead viewing prints.
The focus *should* be at infinity, but the camera's autofocus system didn't actually set it correctly (it front-focused slightly instead). That's really all there is to it.
oheckyeah: Correct me if I'm wrong, but I'm not quite sure the comparison of the center to corner image quality (on the Westminster photo) at wider apertures on the "Image Quality" chapter is fair. At f1.4, the focal plane is so shallow that the "lack of image quality" in the corner may be due to the fact that it's not in the same plane and DOF is causing the blurring.
At that point the Thames is about 250m wide (that's the span of Westminster Bridge, according to Wikipedia). The lens was critically manually focused using 10x live view, so that should be a very close estimate of the focus distance. If you plug the numbers into a depth of field calculator you'll find that the far limit of acceptable focus is about 1500m, even if you use an extremely critical circle of confusion that's equal to the 6D's pixel pitch (~0.006mm). So while the Elizabeth Tower isn't in exactly the same plane as the centre crop, it's not going to be out of focus either.
The take home message from that comparison should actually be just how implausibly sharp the Sigma is at F1.4, even in the extreme corners.
Albert Ang: Andy Westlake / Admin,In page 1 you mentioned the weight is 665 g but on second page it's 815g. Which one is correct? Thanks
The FE 55mm F1.8 benefits from two things: 1) the short back focus of the E mount, and 2) being 2/3 stop slower. This means that it needs to use much less glass (it's a 7 element / 5 group design vs the Sigma's 13 element / 8 group), so can be substantially lighter.
A better Sony comparison would be with the [Carl Zeiss Planar 50mm F1.4 SSM](http://www.dpreview.com/products/sony/lenses/sony_50_1p4_za_ssm), which also has to fit on SLRs, and weighs 518g for an 8 element / 5 group design.
Absolutic: so Andy you are concluding it is the best AF 50mm lens currently on the market. Are you saying it is better than Sony Zeiss 55 FE lens for E mount, that is half the weight, and the price is the same? You mention the 55FE but does not talk about it in your conclusion. I know it is comparing F/1.8 to F/1.4 which is half-stop in difference but nevertheless, which one is better? DXO said at the time that 55FE is the best AF lens they have ever tested.
@Absolutic: Sorry if my response came across as a bit harsh - sometimes it's difficult to separate genuine questions from those simply designed to establish some sort of brand superiority. If you're shooting with an A7R, then I accept it's entirely legitimate to ask which of these has better optics, as you could conceivably use either. But personally I'd choose the Sony FE for practicality.
EvokeEmotion: "We've already published lab test data showing that its optically excellent"
It's, as in 'it is', not its.
Hiring proofreader? I'm in Seattle.
Thanks for pointing that out, I've corrected it. (No need to hire a proofreader when you're so keen to offer your services for free...)
815g is correct. I've corrected page 1, sorry for the confusion. (I've corrected the other error in that spec table, too.)
Dr_Jon: My one concern about this lens is too many people are complaining about AF issues. Is it possible to test it on as many Canon bodies as possible and report what you see? (Canon bodies as you have a Canon version.)
As someone else said, it doesn't matter how sharp it is if it misses focus...
E.g. http://www.the-digital-picture.com/Reviews/Sigma-50mm-f-1.4-DG-HSM-Art-Lens.aspx"The bottom line is that, the longer I focus tested this lens, the less sure I was about its focus accuracy.... (a bit removed as I was over the character limit for a post) ...What I learned from the many hours (actually extended into days) spent shooting and analyzing thousands of images is that the 50 Art lens' AF cannot be completely counted on. Sometimes, most images are properly focused and when my shots counted, this lens delivered. But sometimes, more images are out of focus than I am comfortable with."
I trust Brian to have an objective opinion BTW.
@Dr_Jon: I didn't see any 70D-specific issues, but then again I only shot a couple of hundred images on the 70D.
The variation between reports simply reflects that different reviewers test different copies of the lenses on different camera bodies with different shooting habits. I used two copies of the lens (one in Seattle, another in London), and both were excellent. But a fractionally decentered or tilted lens wouldn't work as well.
@Lassoni: I'd suggest reading up on what the USB Dock offers before drawing conclusions about 'less QC'. It allows both more detailed AF Fine Tune than any camera, and the ability to update the lens's firmware.
Like it or not, SLR PDAF systems normally require some AF fine tuning to match a fast lens to a camera. This isn't about QC, but about the aggregation of tolerances, and the fact that SLR focus sensors aren't directly measuring correct focus at all - just determining a more-or-less accurate proxy for it. It's also worth considering that while high-end SLRs tend to have quite sophisticated AF tuning built-in, entry-level SLRs don't. This doesn't imply that the cheaper cameras have better QC.
thinkfat: This appears to be a very positive review. At least it seems to me, because "positive" is positively used in every second or so sentence.
Regarding the lens, I'm genuinely positive.
I've nothing more really to add than I wrote in the review. I was very confident with the lens's focus accuracy using centre point in the EOS 6D. I wasn't remotely confident it would hit focus perfectly on the 100D. The 70D was somewhere in between. Off-centre AF points seemed less reliable than centre AF points, which tends to be the case with SLRs.
Both of Canon's Live View AF methods, in contrast, were absolutely reliable as always. However the 100D's Hybrid AF isn't really fast enough to be generally useful with non-STM lenses. The 70D's Dual Pixel AF gives decent speed and accuracy, but has other operational flaws.
I'd say that the FE 55/1.8 and the Sigma 50/1.4 are so close that it's impossible to call one 'better' than the other. In practice, you choose between them based on which fits your camera; if you own an SLR, the Sigma will fit and the Sony won't. If you own an Alpha 7, you buy the Sony, as it'll fit directly and focus better. Given this I don't really understand why one should be declared the 'winner'; just accept they're both stunning lenses and be happy they're available for photographers to buy. (Unless of course you're more interested in petty point-scoring than actual photography.)
RuneStenseth: Would it not be better to test those kind of lenses on a Nikon D800e rather than a 22mp Canon? More resolution, dynamic range and no diffusor in front of the bayer array?
@The Otus Eater: Historically Sigma release Nikon mount lenses for testing just as soon as they start shipping them to customers. The Nikon version has just started to go on sale in Japan, so hopefully DxOMark will get one in shortly for testing.
However, I see nothing in the current test data (or indeed the real-world images) to suggest that the Sigma will have any trouble at all keeping up with the FE 55mm F1.8, or indeed the Otus.