Canon EF 24-70mm f/4L IS USM

SLRgear Review
February 3, 2013
by Andrew Alexander

Canon announced the 24-70mm ƒ/4L IS USM in November of 2012 as a lighter, less expensive alternative to the 24-70mm ƒ/2.8. The lens is also equipped with image stabilization, and a specific macro mode.

The lens was designed to fit the 35mm full-frame sensor, and is also compatible with APS-C (1.6x) and APS-H (1.3x) sensors. It also features a constant ƒ/4 aperture across the entire focal range.

The Canon 24-70mm ƒ/4L IS USM ships with a petal-shaped lens hood, takes 77mm filters, and is available now for around $1,500.

Our thanks to Roger at LensRentals.com for supplying us with several of these lenses to test, and for his technical support in analysing the results.

Sharpness
The 24-70mm ƒ/4L IS USM offers sharp results, but is also the subject of some controversy for its performance at 50mm. We tested three copies of this lens; the first and second had poor performance at the 50mm setting. The results you see in this review are from the third and best copy of the lens: some copies will be better than others.

The 24-70mm ƒ/4L has some corner softness issues, which are basically tuned out when the lens is mounted on an APS-C sensor-based camera such as the Canon 7D. There is a very light trace of corner softness when the lens is mounted at ƒ/4, but it's basically tack-sharp from the get-go. That is, except for 50mm - we see it's a bit softer, especially in the top of the frame, than other focal length settings. Stopping down to ƒ/5.6 essentially fixes this problem. From here, it's tack-sharp all the way to ƒ/11, where diffraction limiting sets in slightly, and we note softer results at ƒ/16 and ƒ/22.

The whole story is revealed with the lens mounted on the full-frame 1Ds mkIII. With the aperture set to ƒ/4, we note areas of sharpness in the center of the frame, bounded by significant corner shading, at the 24mm, 35mm and 70mm settings. At 50mm and ƒ/4, we see the same spot of sharpness as on the subframe, but it now covers a much smaller area, and there's significant softness across the rest of the full-frame image area. It's worth noting that these results were found on the third of the three lenses we tested: on the first two lenses, results were poor indeed:

First sample of Canon 24-70mm ƒ/4L IS USM, on Canon 1Ds mkIII, at 50mm, ƒ/4

Stopping down the lens does improve its sharpness: corner softness is never completely removed to provide tack-sharp results from corner to corner on full-frame, but it is dramatically reduced. Peak performance is shown at ƒ/8 across all focal lengths; at ƒ/11, diffraction limiting starts to set in, very slightly impacting on overall image quality. It's more prominently noticeable at ƒ/16, and as usual, shooting fully stopped-down at ƒ/22 is best avoided as we note generalized softness across the frame.

To summarize, it's a good idea to test this lens at 50mm and ƒ/4 during your buying period, to see if the particular sample you're interested in buying performs well.

Chromatic Aberration
Results for chromatic aberration with the 24-70mm ƒ/4L were good: overall, CA is kept under control through the majority of the frame, with only slight fringing in the corners of the image. In this case CA is slightly more prominent at 70mm than at other focal length settings. When CA is apparent, it shows up as X-Y fringing in areas of high contrast.

Shading (''Vignetting'')
Mounted on the sub-frame Canon 7D, corner shading is not an issue at any focal length or aperture.

On the full-frame Canon 1Ds mkIII however, the lens creates images with corners that are significantly darker than the center. In the worst case, 24mm at ƒ/4, the extreme corners are a full stop darker than the center. Other focal lengths at ƒ/4 show corners that are darker to the tune of 1/2 to 3/4 EV. Stopping down to ƒ/5.6 improves this dramatically, however, the only settings you can use which will guarantee you have no corner shading are to use 35mm or longer, and an aperture of ƒ/8 or smaller. At 24mm, there will always be some form of corner shading.

Distortion
Our testing for distortion shows a complicated pattern with a point of near-zero distortion. Used wider than 35mm, we note some typical barrel distortion, not too much in the corners (+0.5% on the 7D, +0.7% on the 1Ds). At 35mm it's still slightly barrel-distorted, but quite near to zero. Above 35mm central distortion stays in the realm of barrel distortion, but quite low, while corner distortion becomes the pincushion variety (-0.1% on the 7D, and -0.4% on the 1Ds, worst at the 70mm setting). All in all, these aren't large numbers, and correctable in post-processing software.

Autofocus Operation
With its USM (ultra-sonic motor) designation, it doesn't get much quieter and faster when focusing the 24-70mm ƒ/4L IS USM lens. Going from infinity to close-focus takes well less than one second, and point-to-point focusing is lightning quick. Focusing results can be adjusted at any time by just turning the focus ring, and focusing operations do not rotate attached 77mm filters.

Macro
The Canon 24-70mm ƒ/4L IS USM has some special provisions for macro work. A special macro mode is included that is activated by pushing the macro switch into the ''macro'' position and turning the zoom ring to the macro setting: this allows the lens to offer a minimum close-focusing distance of just under eight inches: with the length of the lens considered (3.7''), this gives a working distance of around four inches from the front element. This means that, unless you have macro lighting, the lens itself might shadow your subject. That said, in this configuration the lens offers an impressive 0.7x magnification. There is also some slight fine-tuning adjustment available for zoom and focus in the macro setting.

Build Quality and Handling
As an L-class lens, Canon doesn't spare much in the way of expense to ensure that these lenses keep going and going. The 24-70mm ƒ/4L IS USM is weather- and water-resistant, employing several rubber gasket seals. The front element employs a fluorine coating to reduce fingerprints and smears; the design of the lens employs 2 aspherical and 2 UD lens elements, and an aperture composed of nine curved blades.

There are three switches on the lens: one switch to enable or disable autofocus on the lens ("AF / MF"), a second switch to lock the zoom to the 24mm setting and activate the macro mode, and a third switch to activate and deactivate image stabilization. A window provides distance information in feet and meters: there is no depth-of-field scale, but there is there an infrared index at the 24mm and 35mm focal length settings.

The focusing ring is composed of rubber with small ribs about a half-inch wide. The ring provides excellent manual focusing fidelity, with a slightly smooth resistance and plenty of travel. The ring ends in soft stops on both ends of the focusing spectrum, and will focus past infinity.

The zoom ring is composed of rubber with large ribs, also about a half-inch wide. The ring takes about ninety degrees of turning action to go from 24mm to 70mm, and again, the ring is very smooth, firm but not too tight; it requires two fingers to move. Zoom creep wasn't a problem with the lens we tested, but if it did become so, there is an option to lock the zoom ring at 24mm. There is some lens extension when the lens is zoomed out to 70mm; the lens grows by 1 1/2 inches.

The EW-83L lens hood is petal-shaped, attaches via a bayonet mount, and adds about 1 1/2 inches to the overall length of the lens. The interior of the hood is flocked to reduce any stray light, and there's a locking mechanism to keep the hood attached to the lens. It's also worth noting that Canon includes a new lens cap with the 24-70mm ƒ/4, which employs a front-pinch design and is much easier to use.

Unfortunately, we didn't have enough time to conduct an image stabilization test with this lens, but we hope to have it back in the future to do so.

Alternatives

Canon EF 24-70mm ƒ/2.8L II USM ~$2,400
There's a reason you pay almost $1,000 more for the 24-70mm ƒ/2.8: in addition to the faster constant aperture, its image quality is much better, and much more consistent, than the 24-70mm ƒ/4. It doesn't have image stabilization, however.

Canon EF 24-105mm ƒ/4L IS USM ~$1,150
Less expensive and offering the same L-class build quality, the 24-105mm offers image stabilization and a longer telephoto focal length, but doesn't have the macro capability and is slightly more distorted in the wide angle. In terms of image quality, the 24-105mm is the same or better than the 24-70mm, especially at the 50mm setting.

Sigma 24-70mm ƒ/2.8 EX DG HSM ~$900
The Sigma 24-70mm ƒ/2.8 is definitely much less expensive than the Canon 24-70mm ƒ/4, but this is definitely a case of getting what you pay for; in addition to the better build quality of the Canon L-glass, the corners are much better for sharpness in the Canon than in the Sigma. However, the Sigma doesn't have the same inconsistency at 50mm that plagues the Canon.

Tamron 24-70mm ƒ/2.8 Di VC USD SP ~$1,300
We haven't yet tested the Tamron 24-70mm, but it does offer image stabilization and the larger ƒ/2.8 constant aperture, at around the same price point.

Conclusion
Canon made a smart decision to offer a lower-priced alternative to its 24-70mm ƒ/2.8, in order to give Canon shooters of the more casual or economical nature a way to stay true to the brand name. Unfortunately the lens' performance at 50mm stands out as unacceptable, especially when you consider the lower-priced option here still costs $1,500. Coupled with the fact that we had to seek out the best version to get even these results, this does not breed confidence in the lens.

So what is a Canon shooter to do? If you're happy with the performance you see from our sample pictures, and you need or want the image stabilization and macro capabilities of the 24-70mm ƒ/4, then you should be happy with your purchase - assuming you get as good a copy as we did, and not something worse. If you demand absolute image quality in this range of focal lengths, you'll have to be prepared to invest in the 24-70mm ƒ/2.8L II USM. For everyone else, I don't see a compelling reason not to go to the tried-and-true 24-105mm ƒ/4L IS USM. It offers image stabilization, a longer telephoto focal length, and similar but more consistent optical performance, all for a lower price tag.

Product Photos

Sample Photos

The VFA target should give you a good idea of sharpness in the center and corners, as well as some idea of the extent of barrel or pincushion distortion and chromatic aberration, while the Still Life subject may help in judging contrast and color. We shoot both images using the default JPEG settings and manual white balance of our test bodies, so the images should be quite consistent from lens to lens.

Still Life shot

VFA target

As appropriate, we shoot these with both full-frame and sub-frame bodies, at a range of focal lengths, and at both maximum aperture and ƒ/8. For the ''VFA'' target (the viewfinder accuracy target from Imaging Resource), we also provide sample crops from the center and upper-left corner of each shot, so you can quickly get a sense of relative sharpness, without having to download and inspect the full-res images. To avoid space limitations with the layout of our review pages, indexes to the test shots launch in separate windows.