As we saw in the previous Optical Performance Shootouts, the 16-35 F4 and the 17-40, the 16-35 F4 consistently delivers critically sharp images throughout the entire frame, whereas the 17-40 produces marginally softer images in the corners, and with a slight amount of CA present when extreme focal lengths and F-numbers are used in conjunction.
Canon lenses have historically performed very well in the center, but always seem to have a bit more difficulty on the corners. Not so with the 16-35 F4. Center sharpness is almost as fantastically critically sharp on the corners as it is in the center. Impressive!
Now let’s get out of the testing cave and take a look at landscape results taken in the real world.
The Olympic Mountain Sunset is quite a sight, and yet capturing it can be difficult due to the extreme dynamic range above the mountain range and the foreground shadows. In this long exposure I used a 6-stop circular ND to stop the light down and push the exposure out, and a 3-stop Hard Reverse GND to balance the dynamic range, and a 2-stop soft to soften the hard line of the 3-stop Hard Reverse GND – so 6 + 3 + 2 stops to bring the dynamic range of the A7R from 14-stops closer to the scenes 20+ stops of dynamic range.
Click here to download the RAW and TIFF image
24mm is one of my favorite focal lengths on both the 16-35 F4 and the 17-40, and based on the results I’m seeing from the 16-35 F4 it certainly resolves critically sharp at 24mm, especially between F14 – F20. But then again tack-sharp images from the 17-40 is also commonplace, so nothing exciting here, until you look at the distant ridge lines – zero chromatic aberration. The 17-40 always has a small bit of CA on sharp edges with dramatic dynamic range differences – not so on the 16-35 F4.
Regarding the function of the lens in the field, the threshold for IN FOCUS and OUT OF FOCUS is much smaller than other lenses, 17-40 included. That means that it’s easier, and faster, to get the lens in sharp focus.
Just how sharp is critically sharp? Take a look at this 100% close-up from a 229 second long exposure at F18, ISO 64 at 16mm on the A7R’s 36MP sensor:
Even the smallest details in this image render tack sharp. It’s easy to muddy the details on the foreground subject in this image, however the 16-35 F4 successfully nails this one. Download the hi-res 216MB TIFF here to see for yourself!
In nearly all my tests the 16-35 F4 performed critically sharp, and when shot on my Sony A7R it’s still tack sharp down to every pixel, even despite nearly double the pixel count of my 5D3 and 6D. If Canon is getting ready to launch high resolution sensors in the near future, this lens proves that it’s ready today.
Click here to download the RAW and TIFF image
Sharpness throughout the frame is impressive, and starbursts are well pointed.
The foreground details, although very dark, are not muddy and they’re tack-sharp where the dark area meets the water.
In this 366 second long-exposure landscape photograph the Canon 16-35 F4 captured it nearly tack-sharp all the way down at F22. More impressively, however, is the lack of CA on an incredibly high dynamic range area. It’s all too common to have thick red lines bordering this area of the image on other lenses, wide-angle or not.
Sharpness throughout the frame is impressive, from the bird on the distant rock in the top left, to the corner sharpness in the lower right. Zero CA on the mid rock on the horizon. Overall it’s an incredibly clean file.
Chromatic aberration (aka color fringing) on the 16-35 F4 is incredibly well controlled, and slight CA can be found in scenes which have extreme dynamic range at 16mm up to about 24mm – but it’s rare. Beyond 24mm there’s virtually zero noticeable chromatic aberration when magnified at 36MP.
When composing in Live View on a Canon or Sony at 100% there may be noticeable CA in the corners of the image on the LCD or in the EVF, however this is due to the lens being stopped down to it’s lowest F-number (F4 in this case) for the purpose of composing, but when the shutter actuates the result will likely not have any at all. Beyond F8 I haven’t seen any noticeable traces of CA until F20 to F22.
Sure, CA can be corrected in post-processing quite easily, however lens correction will progressively soften the image the more you reduce CA. Less lens CA ensures a sharper image in post-processing, because ideally you don’t have to apply any CA correction under lens profile corrections.
To get a better idea of CA, take a look at this 100% close-up above. In high dynamic range scenes such as this I would expect to see color fringing occurring on the edges of the rocks where the dark detail-less part of the rock meets the sky. The 17-40 sometimes produces a thin line of CA in these scenarios, whereas the 16-35 F4 has none. None at all. To download the hi-res TIFF of the below image for reviewing 100% sharpness and CA, click here.
CHROMATIC ABERRATION SUMMARY
The 16-35 F4 has the best CA performance on any wide-angle lens – prime or zoom – I’ve ever used. The 16-35 F4 impressively outperforms the Canon 17-40 with regards to CA.
I’m seeing slight light falloff (corner darkening) of 1-2 stops at F4 at 16mm – without a filter. Even 1mm makes a significant difference at the wide end of the focal range, and vignetting therefore becomes more pronounced. In my testing I found UV filters with a frame thickness of 4mm add an additional 1-stop of light falloff, pushing the total light falloff to somewhere around 2 or 3-stops.
With filters which are 4mm or higher, I’m noticing vignetting in upwards of 3-4 stops at 16mm. The 3.2mm and 3.5mm UV filters that I’m using don’t have any noticeable vignetting above what the lens itself already produces.
To avoid compounding the already inherent 1-2 stops of vignetting found in this lens, use only UV filters with a frame thickness of less than 4mm, ideally 3.5mm or less.
All of this said, when profile corrections in Lightroom are applied it will remove distortion and 90% of the time this entirely eliminates the slight vignetting that occurs when shooting without a filter.
VIGNETTING / LIGHT FALLOFF SUMMARY
The 16-35 F4 has inherently darker corners when shooting at 16mm at F4. Using filters makes matter worse, especially if the thickness exceeds 4mm.
The 16-35 F4 is considered an ultra-wide angle zoom, so distortion is expected (and sometimes welcomed!) with lenses in this focal range.
Depending on the number of subjects with straight lines in your composition, the difference between 16mm and 17mm can be quite significant. And to be honest, I much prefer 1mm at the low end of the focal range as opposed to 5mm at the high end.
Barrel distortion becomes more pronounced the wider you go, with 16mm having the most visible effect. At 24mm distortion is nearly eliminated and profile corrections cease to have much visible effect. At 30mm the lens starts to pincushion, becoming most pronounced at 35mm.
The lens profile corrections for the Canon 16-35 F4 is already present in Lightroom and other applications, so distortion can largely be controlled in post-processing. Yet at 16mm compositions with inherently straight features, such as a horizon, may still appear distorted between 16mm to 20mm or so. Go to the top of this page and view the vertical Golden Gate Sunrise and notice how the bridge leans ever so slightly to the right? That’s the effect at 16mm, even with correction applied.
It’s an ultra-wide angle lens – it’s gonna have distortion
Flare and ghosting is well controlled on the 16-35 F4. When flare and ghosting is present contrast suffers, so color is better as a result of having better contrast.
I love using classic Nikkor and Canon lenses, yet one of the main problems with those lenses, although incredibly sharp, is that they tend to flare often and with weird and even normal light sources, not just the sun. When the lens flares it lowers contrast and as a result lowers color saturation. The high-tech coatings applied to modern optics reduce such optical phenomenon. All of the jargon that Canon uses (which I won’t recycle in this review to make myself seem smarter) is responsible for reducing flare and ghosting.
Compared to the 17-40, the 16-35 F4 exhibits far less flaring and ghosting. Most of the flaring that I experienced with this lens was shot at 16mm at F16 to F22 with the sun in the corner of the frame, directly hitting the lens. But aside from the sun directly in the frame at 16mm, I’ve noticed just a slight flaring at other focal lengths in the 16mm to 24mm range.
FLARE & GHOSTING SUMMARY
The 16-35 F4 is best in class for flare and ghosting.
Like this article? I publish content once a month or so. Click here to join my email newsletter and stay up-to-date!