I have wanted the Sigma 35mm f1.4, since it was announced a little over a year ago. Having other things to spend my money on, I held back from the purchase at the time. Fast forward to December 2013, I decided to revisit this "Art Series" lens for street photography - the 35mm and 50mm FOV are perfect for my style of shooting.
Build & Aesthetics
Overall, this 35mm is beautifully crafted and it looks really pricey, even though it only costed me $899 AUD.
The exterior of the 35mm is just gorgeous with a silky black finish that is not prone to fingerprints, and the manual focus ring is wide enough that no matter where my hand is on the lens, I am able to grab onto the focus ring instinctively and quickly.
After playing with it for a while, I started to find it to be a little too heavy at times to be used as a walk-around lens. The thing is beefy and weighing almost as much as my Nikkor 24-70mm 2.8. But when I nailed shots with this lens (especially up close to the subjects), the results are almost orgasmic.
Sharpness & Contrast
And I have to admit, I do have a love/hate relationship with this lens. When good focus is achieved, the results are nothing short of amazing - even when shot wide open. I have yet to see lenses which are this fast, and that produce this level of sharp and contrasty details at the point of focus. Normally, lenses have to be stopped down a couple in order to resolve details this well. 5/5 stars from me in this department.
The Autofocus Issue
I have read about this issue on a wide range of camera gear forums and also independent reviews from other photographers - this lens may have issues with its AF - if you are unlucky. Sad to say, I lucked out, twice.
Outer Points AF
My first copy of the lens was tack sharp at f1.4 within 6-7 ft. Focus on anything beyond that however and getting the AF to hit its mark can be really frustrating. The center AF point acquires autofocus correctly at this distance about 60% of the time while the outer AF points nail focus around 10% of the time. The further out I go on my D600, the worse it gets - which is understandable since the cross type AF points are all clustered towards the middle of the frame on the D600.
The question is, why does it rarely happen with my other Nikkor lenses? I've been comparing this with my Nikkor 50mm 1.8G (a dirt cheap plastic body lens) and the Nikkor rarely ever misses.
A Better Second Copy
I took the first copy back to the store and explained my situation to the sales staff, and after a few test shots, they confirmed the issue and invited me to try out their other copies.
After going through 5 other copies at the store, I finally found one that neither back focuses nor front focuses as often (the rest of the copies I tested there were just as bad as the first one). It was actually hitting the mark most of the time while I was testing it in the store (the store was lit predominantly with fluorescent and tungsten lamps). Note however, that since the store was rather small, I didn't get a real chance to test it at long distance focusing.
After taking the lens home, I decided to take it out onto the streets to test out how well the lens performs under real world shooting conditions. I made note to only shoot with the center AF point and not to lock focus and recompose for this test - to see how the most sensitive AF point on the camera handles focus with the Sigma.
The funny thing is, under shade on a bright sunny day, the autofocus gets even worse. I ended up missing 70% of my shots at the widest aperture. I didn't realise it at the time of shooting looking at the back of the camera, until I took it home and uploaded the shots to Lightroom 5. Most of the shots were slightly soft, where the front or the back of the subject was tack sharp (this was very random as well and not consistent enough for micro AF adjustments).
The Sigma USB Dock
I purchased Sigma's lens calibration and firmware upgrade dock, which provides users with 4 distance categories to fine tune the AF. In my case, I only had to concern myself with the 6 ft to infinity focus range.
In general, lenses either front focus or back focus and if this is happening consistently, micro adjustments can be fine tuned within the camera body. The problem with this Sigma 35mm is that nothing is consistent when focused at 6 ft to infinity. I spent days tweaking the AF only to drive myself nuts in the end.
I even printed out focus charts with back/front focus rulers, and performed the tests on a heavy duty Manfrotto studio tripod, and even lit the charts up well with constant lights in the studio. The test shots at the long range of the lens were in focus but never 100% sharp no matter how I fine tuned using the USB dock.
I then took it out and shot it in real world scenarios and the AF was all over the place. All the tests were done outdoors during daylight mind you. Sometimes it focused behind the subject (the AF actually focused on the background about 4 metres away) and sometimes it front-focused by a few metres. Again, let me reiterate, this has never happened with any of my lenses - they don't front or back focus by that much.
Bottom line is, the dock is useless to me with this copy of the lens and I have spent way too much time tweaking rather than enjoying photography.
This lens is dangerous and too unreliable for use. Sure, it looks great when shots are in focus but if I can't trust the AF, the lens is useless to me. No lens is perfect but I'd rather have a slower lens that nailed focus most of the time than a fast 1.4 that is erratic with its focusing. With a f1.4 lens, I'd like to shoot it wide open most of the time so manual focusing is out of the question - it just too difficult unless you have a focus peaking feature (yet to be seen on today's DSLR with optical view finders).
The problem with the infinity focusing is not really a back or front focus problem, it is an autofocusing consistently issue. With other lenses, if the camera fails to acquire a point of focus, it normally hunts, tries and fails. With the Sigma 35mm, it always confirms focus regardless, which I think is very misleading to the user.
My copy of the lens will go back to the store tomorrow for a full refund. I've had enough of testing, though I might revisit the it again 1 year from now, when Sigma has nutted out all the issues. The "Made In Japan" tag on the lens just doesn't mean anything these days.