New samples from the Sigma 24mm F1.4 DG HSM Art lens

Sriously?! I own the Sigma 35mm Art and 50mm Art and use them professionally; they are both far more reliable in terms of autofocus than any of the other (more expensive) Nikon lenses I've owned in the 24-50mm categories. The 35mm Art focuses so reliably that I literally never miss a shot. Both are ridiculously sharp. I think you must just have a (rare) bad copy as most reviews have rated both the 35 and the 50 very highly.

Mine works sometimes and then it stops focusing at all. I have to make it blurs first and then it will focus again. It is weird. And I have it on a full frame. Artysta, what body do you use?

I was about to consider the 35mm sigma Art until I read some reviews complaining about the autofocus system so I chose to avoid the lens.

The problem seems to be Nikon and Canon's patents preventing other manufacturers from using autofocus correctly on their cameras. Which is why none of the carl zeiss lenses for Nikon or Canon autofocus.

I remember carl zeiss saying something about that a few years ago, how ridiculous the situation is and how other manufacturers are simply shut off from having proper autofocus. I see lens manfucturers focus more and more on other brands that seem to be have less patent problems, including Sigma and carl zeiss.

Also note, that none of the lenses Sigma / Tamron or Carl Zeiss make for other OEM, other than Nikon or Canon, seem to suffer from autofocus problems, it is always a Nikon and Canon problem, according to carl zeiss, because of the unwillingness of them to open up their patents.

Have both the S35A and the S50A and use them on eos 5D2, 5D3, 7D, 450D and have no focusing problems whatsoever. The microadjustment via the USB dock worked like charm. For unfortunate customers having a bad copy I would recommend to return them or let them be checked (happend to me with my EF14/2.8II and the now gone EF50/1.4).

I own the 50mm Art, and while it does miss occasionally, I find it isn't any worse than Canon lenses I've owned. I do find it has trouble focussing on subjects without points of contrast(areas without color difference), but you will have that trouble with any lens.

... which allows you to fine tune a lens even better than the microadjustment values on Canon / Nikon bodies. In fact, I believe it was once stated somewhere that Canon suggests you microadjust your lens with a subject at 50x focal length, or 25x focal length if you tend to shoot subjects closer with that particular lens.

But that only optimizes the lens for that particular subject distance. For example, I'll tend to shoot subjects between 10-25x focal length w/ a 35mm F1.4 prime, & since DOF is even shallower at 10x than 25x, I microadjusted my Canon 35L w/ a LensAlign at 10x focal length (350mm). But now infinity focus is way off, & any subject I'd shoot far away (e.g. >50x focal length) would have a terrible purple fringe around it.

Sigma's dock actually allows me to get around that, so I actually prefer Sigma's 35mm Art over similar offerings from Canon & Nikon.

Did you try microadjusting your Sigma? If not, I don't think you can claim it didn't focus 'reliably'.

"The USB dock is an admission of this, and represents a partial, duct-tapish solution."

That's like saying that any product that is firmware/software upgradeable is an admission of fault or weakness! How terribly short-sighted. In the modern era, products should be upgradeable through firmware/software updates. It doesn't matter if you're talking about smartphones, cameras, lenses, desktop software, or even automobiles! There should be some mechanism by which updates can easily be installed in order to improve performance, add features, or simply respond to unforeseen changes that didn't exist at the time of a product's introduction!

Rishi,

This is a bit too much. I own the 35L, and just took two consecutive shots at 30cm or so - the closest I could focus. Perfectly focused. I use it all the time focused much farther away, form 2-3m to infinity, and I know it focuses perfectly. It just does.

I had only one Canon lens that needed different MA at different distances. It went to Canon and came back adjusted. It was defective, as simple as that.

The Sigma 85 was terrible in that respect. It was unusable.

"CaNikons"?

... and I've had a Canon 24L and 85L that both needed AF micro-adjustment (AFMA) values in the teens, in opposite directions, on my 5D Mark III...

This is a reality of traditional PDAF systems. If you haven't had to microadjust (and are absolutely sure you didn't need to - as in you did the correct, controlled tests to absolutely ensure that you didn't need any more AFMA than the person complaining about the Sigma), then you're simply lucky.

The chances that your body & lens, that went through different facilities, different calibration apparatus, are perfectly matched are... well... not zero, but not 100% either.

All I'm saying is that it can't be generalized that one brand's lens requires AFMA more than another brand's. It might, if there are serious issues (actually, that Sigma 85mm had serious residual spherical aberration IIRC, & that could cause issues), but it can be generalized. Neither can the need for AFMA at diff subject distances, for diff lenses, be generalized.

These images were taken in a couple of hours and I was lucky to be able to take them at all on a pre-production lens.

They should consider, period. My absolute favorite lens is 35 mm f/2 and I shoot it open almost all the time. There is nothing to go wider and stay open. The Pentax 21 mm is an excellent lens, but it is only f/3.5. The Sigma 24 mm f/1.4 makes perfect sense on APSC. And Pentax is a much bigger market than Sigma expects, as there are so many Pentaxians shooting with prime lenses! I wonder if Pentax User base * Prime lens users > Canon User Base * Prime lens users? So the Pentax lens market for prime lenses is actually bigger than the Canon one!

My guess is that since very few of the 35mm f1.4 Art lenses sold in Pentax mount, Sigma decided not to make the rest of their full-frame Art series lenses in Pentax mount. How do I know this? Well, by the number of reviews at B&H it is indicated. There are just three reviews of the Pentax version of the 35mm f1.4 A, but there are 103 reviews of the Nikon version and 213 reviews of the Canon version. When Pentax actually sells a full-frame camera, they should pay Sigma to make all their current lenses in Pentax mount. I doubt they will do that though, because that might eat into their own lens sales. It will be a catch 22 for them. Do they dish out good money to get a competitor to support them and possibly sell more full-frame cameras, or do they leave it alone and possibly sell the same number of lenses anyway?

Scottelly, I could see the 35 not selling so well in Pentax because now there is some comparable AF competition at 35 (FA 31 f1.8 Ltd, FA35 f2, DA35 f2.4). However for 24mm this is not as much competition (second hand FA* 24 f2 or MF Samyang 24). The DA 21 is nice and small but slow, wider than 24, and of course not FF compatible.
So I think the Sigma 24mm could sell well in Pentax mount.

Expecting symmetry would be a major error in logic. That would require each lens to be assembled perfectly, which is not realistic in the world of today's MBA-driven consumer electronics and optical companies. Nowadays symmetrical results should be classified in the realm of "happy surprise."

Actually, LR demonstrated that the differences are measurable but hardly visible except in a minority of cases.

You're crazy. Don't put your ego in it. People can have an opinion on bokeh (which is a highly subjective thing to begin with), while simultaneously admiring this lens for its sharpness at close-distances from the examples.

The reason no one is talking about other measurable things, is because no one has measured them yet. The one thing we can easily see is bokeh, because these are mostly shot wide open.

Talking about bokeh of a fast lens! How weird!

Was the OP trashing Sigma? I thought his post was directing sarcasm at Canon/Nikon fans who cannot admit that a Sigma lens is superior in any way.

What is a big deal about the sharpness of any lens these days is the 'across the frame' sharpness at wide apertures. For example, I can shoot the same infinity scene with my 24-70 II at f5.6 that used to require f11 with many lenses, zoom and prime, from previous eras. Why do I want to avoid f11 when I do not need it for legit DoF purposes? Diffraction.

Well it depends on what you want to achieve. Sharpness is just one aspect of image creation. However it is much harder to create sharpness then it is to reduce it. My experience is better to have then not. Also the increase in MP of digital camera tests the resolving power of a lens. If your lens cannot meet the test then you are essentially collecting useless data. Stopping down a lens has limitation most lenses reach optimal sharpness 2 stop down from maximum aperture and they resolving power decreases with smaller apertures. If you must use DOF to maintain focus you will be again benefit from a sharper lens as the loss begins from a higher resolving power. Often though careful focusing (a lost art) can achieve what is needed at larger apertures. So yes diffraction, but again its intuitively knowing what you want to achieve then making choices based on equipment limitations or at times leveraging the limitations to your benefit.

Here we go again, more posters claiming to know what someone else needs.

If YOU can't see a reason to purchase a lens or a camera then don't.

If YOU think the lens is to expensive then it most likely is for you so don't buy it.

But don't tell OTHERS that they don't need it for the type of photography they do.

Do tell others about YOUR experiences with a lens you own as all of us are interested in real life experiences.

I use the Sigma 50 Art when called for to shoot landscapes with a Nikon 800E using a tripod to support the creation of a sharp image and own both the new Sigma 50 Art and the little Nikon 50f1.8G and I mount the little Nikon as a walk-a-round because of its small size and weight and find it okay if I don't crop in and print large, but your experiences may differ.

Again, if you can't see a reason or a need for something or in you mind its just hype, then don't buy.

Yeah! ) At least that's better than nothing.

When I want out of focus circles, the first thing I reach for is a 24mm lens. Unless I have an 18mm available.

Way I see it - 'boring' is exactly what bokeh should be. Free of color fringes, 'nervousness', etc. 'Boring', so the main subject (in focus) can pop out from the boring bokeh, rather than be distracted by less-boring bokeh calling attention to itself.

So maybe he was trying to say it in a positive light.

Maybe not.

Hence my request for clarification.

"I will probably text a few copies to make sure I have a good one"

Old myth. It's 2015, not 2005. Besides, this is their top line.

Not really. I see copy variation quite frequently.

More likely, you see people complaining about it ON THIS WEBSITE, even if there are no issues at all. They just love to go to shops and exchange a lens, to feel important. You know, like those people who pretend to know wine and send bottles away because they allegedly aren't up to their standards. The same people who regularly get fooled by cheap wine smuggled as top-notch...

No, I mean I buy multiple copies of lenses, we get multiple copies of lenses sometimes, and I see copy variation between them. We even replace studio lenses from time to time because they go bad.

I don't think it's a non-existent problem.

Not saying sample variation is not existent, but according to some people here, Sigma sells absolutely atrocious besides absolutely wonderful lenses. And it can't be a coincidence that I've never experienced a single bad Sigma lens since I started buying that brand...20 years ago maybe?

So yeah, sample variation happens (with ALL brands, not just Sigma), but some people here wildly exaggerate the extent of it.

Roger Cicala has some data Lensrentals that shows variations on some lenses.
Certainly not a myth.

Exactly. Although I bet it's getting better with tighter tolerances today.

Copy variation is precisely why it's difficult to even begin thinking about AF accuracy/precision tests... as poor optically performing copies will also affect AF performance. A bit of a tangent, but your mention of Roger's excellent work just reminded me...

I agree. I think my D810 would give sharper image performance. At this early stage, I really can't be sure. But, judging from the 35 & 50 ART lenses, I have to surmise that the new 24 will eventually show the same sharp images. These early examples probably should not have been posted, as they are not really lab type conditions like DXO conducts. But, I applaud Dpreview for their apparent "scoop" of the rest of the industry.

"The samples were shot on a Canon 6D which doesn't have low pass AA filter."

False.

joe85:

There's a low pass filter listed in these specs:

http://www.adorama.com/ICA6DZ.html?adver=new&

Yes, it does have a Fluorinlayered low-pass filter. I think that Joe85 actually meant to say that; but mispoke himself. That was why I responded by mentioning my D810 (which does not); meaning that, on my camera, the images might be sharper. No harm done. I knew what he meant.

Thing is, I don't believe that story and its clear why Sigma wouldn't believe it.

If that lens had a production defect so abd "the end just fell off on its own", it would have done so looong before you ever got your hands on the lens. In shipping and transport.

It just happening on its own sounds like a typical bull excuse story.

(...)If that lens had a production defect so abd "the end just fell off on its own", it would have done so looong before you ever got your hands on the lens. In shipping and transport.
It just happening on its own sounds like a typical bull excuse story.(...)
This is just great !

Hello Lt Colombo,
So, you're building your own little theories about what really should append, (hey of course, a lens can't broke during camera operation), and claim in public than i'm a liar.

thank you :)

This is an absurd limitation to adopt for yourself. Lee Jeffries shoots haunting street portraits using exclusively a Canon 24 1.4L on a Canon 5D. http://leejeffries.500px.com/homeless

It's not absurd at all. Getting in close to a face is great, but doing so with a wide angle can distort features if you aren't careful. Look at the portraits that StevenE linked to. Sure they are in close, but notice how they either directly face the person, or are at a right angle profile.

The lens is a tool, and if you are going for a certain look then by all means use it. You just have to be aware of its behavior in certain shooting situations, that's all.

oh sorry - didn't know
that some of you don't know physics...
If you go close then 1 m to a face with a wide-angle or even with a 50 mm or 60 mm - you will see a heavy distorted face -
see all the dump IPhone portraits!
The best lens to show a face naturally as it is - will be a 90 mm (or 85 mm; or 100 mm).
You can take a 50 mm - when you don't move to close - but a 24 mm is only for trying some strange art with faces or great for stretching the hole person.

QuarryCat ... spare me the internet-expert crap.... a good photographer will use the appropriate lens for the effect he/she wants.
Only a sanctimonious ass will insist on an 80 - 135 mm for all portraits. You stick to your rules to impress newbies while others, like Lee Jeffries, deservedly win awards.

I'm just glad that cinematographers & photographers like the award-winning Emmanuel Lubezki don't ascribe to such limiting views with respect to portraits in their photography.

while one, two people with camera use a wide angle - and then crop the face out - and make some unusual face pictures...

you can't be serious to take a 24 mm for face pictures - crazy.

if you want to do effects - go on and leave me doing photos.

it is a different thing to crop out of a wide-angle or 50 mm - then to go as close - as you need - to get the full face.
If you don't get that - I can't help.

If you don't think about nature rules - your pictures from people will look very ugly.

... and Emmanuel Lubezki just won a 2nd Oscar, shooting faces with wide-angle lenses.

"E. Lubezki for sure knows what to do with faces... but he is filming, I spoke about full frame photography - and not cinemascope."

You did take a look at his wide angle portraits on his instagram page, right?

You do realize there's an entire type of photography called 'environmental portraiture', dedicated to wide-angle pictures of, well, faces, right?

You do realize there are actual types of photography apart from shooting your tigers, right?

Because your comments would appear to indicate you are not aware of any of these things.

^You want cartoon glasses, a fake nose, and mustache too..?

Its goofy. "Distracting." Like Goofy. You ever watch Disney as a kid? I was trying to politely not say "bad," which is what I really should have said, but if I'm objective, its actually acceptable, so I'll settle for the word "goofy," instead. Someone else can call it "bad;" I'm not there yet. But in some cases (not all), it does distract from the rest of the image. You don't need arrows, the bokeh practically calls you to itself.

Also, I think I should modify my response -- its not just the far-range bokeh, its all the bokeh. It just tends not to be as pronounced on near objects and the closer it is to the focus point, but gets goofier the further away you get.

Have you ever looked at the bokeh from other 24mm AF lenses? The direct competition is even goofier.

That is one *very* overcast day. Not sold.

^Sounds great. I bought the 50/1.4 ART recently for Nikon, and its a fantastic lens, couldn't return it. Probably the sharpest in my bag now, next to my copy of the Nikon 85/1.4G.

I get a lot of "micro-hunting" I'd call it when focusing in general with this lens (similar to the Nikon 85/1.4G), which is a bit more than other lenses, but generally not bad or far off. Continuous mode made it worse on the D700, but I can't see much of a difference on the D810, which somehow seems to do better. Single focus is obviously more confident, but still gets a bit of a jitter. As far as tracking and moving subjects, I'm not feeling that is this lenses forte, as it goes out of focus fast. And I may not be understanding the technical behaviours I'm seeing from an engineering perspective to comment. Basically, I wouldn't use this lens when walking towards something, haha.

We are different "we". :=)

I still hope that Canon can fix the egg-to-hard issue with a firmware patch.

well there is something called an in depth analysis where we can also compare how much better the lens is and in which areas. as well as compare the number of new features etc
Hope that answers your question about whats to compare

@ragmanjin - not if you're shooting moving subjects :)

Even with static subjects it's hard. You have to zoom in to the image in Live View, manually focus, then zoom back out to ensure composition / framing - at which point you or your subject may have moved enough to shift the focus by the time you depress the shutter.

At 24mm F1.4 and close subject distances, your simple rocking back-and-forth on your feet is enough to change the plane of focus, especially on a high-resolution sensor. That's exactly what happened with some of the shots I took w/ the 24/1.4 on the A7R.

But, yeah, otherwise the Samyang looks great. I do wish DxO tested axial CA though... that's the tougher one to remove.

Fair, but I do think it's worth noting that more than 100 years worth of society-altering photos were taken with manual-focus lenses, without the aid of live view or image stabilization. I wonder, how many rounds of the Olympics would that be?
I don't mean to sound cynical, AF is very convenient. But it's by no means the be-all and end-all of photography, especially with well-made MF lenses and their focus distance scales and damped long-throw rings.

Modern DSLR's are not designed for manual focusing - old camera were.

"Fair, but I do think it's worth noting that more than 100 years worth of society-altering photos were taken with manual-focus lenses, without the aid of live view or image stabilization. I wonder, how many rounds of the Olympics would that be?"

The other guy said it - older cameras literally had fantastic aides to help you focus. Newer DSLRs don't.

Furthermore, few were scrutinizing (pixel-peeping) with as much ease as we pixel-peep now, due to the easy & availability of high resolution displays, high resolution printing devices, etc.

So it's a totally different ballgame. The point is, standards change, and unless you're holding everything to the same standards, there are technological advancements that can help you get performance far ahead of what manual focus will get you, for certain scenarios, today.

Honestly I'm not sure how focus confirmation is handled in Canon or Nikon mounts. I do know Pentax has "catch in focus" if you enable it, meaning the camera fires the shutter as soon as the image is in focus (there are a lot of manual focus lenses for the K-mount, they have certain features like that for their vintage glass). So while can be very difficult to use MF lenses with lower-end Canon/Nikon bodies using their tiny pentamirror viewfinders and fixed focus screens, there are many MF aides out there if you need them, either as add-on accessories or included features in certain brands. I hear Fuji and Sony mirrorless EVFs are great for that as well.
I'm not trying to say one way is better than the other, but rather that there's no need to write off manual focus completely when some of those lenses (i.e. Samyang, Zeiss, old Takumar primes, etc.) are still so thoroughly good if you have the patience to use them.

Yup, just goes to stress the importance of AF when shooting fast lenses. Between the time I'd focused the 24 Art on the A7R, either the subject or I would typically have moved enough to throw focus off with the razor-thin focus of 24/1.4 & close subjects. Although every now and then, it'd turn out fine, serendipitously.

Also, remember that 36MP shows focus errors more than 20MP. Axial CA on the 6D also looks much better than on the A7R - b/c the lower resolution sensor masks it to a certain extent.

@Glassfish - can't comment yet on AF, but if it's anything like the 35mm F1.4 Art, it'll be great & fast. Which is impressive given how much glass has to be moved.

Btw, any of you guys have experience with AF tracking on the 50mm Art? We've been having issues with ours, that we don't have with any of the 35mm Art copies, and we're wondering if it's just a sample thing or whether it's more widespread.

Yes, Samyang has a 24 1.4 that covers the full frame image circle.