Studio Tests - 35mm full frame
The 70-200mm shows its true colours on its native format, with excellent corner-to-corner sharpness at most focal lengths and apertures. As might be expected, it's weakest towards the telephoto end, or when shot wide open at any aperture, but even here performance is none too shabby.
|Resolution||Performance on full-frame is very good; the lens is a little soft wide open, and sharpness drops off a little towards the long end, but in its 'sweet spot' the lens is impressively sharp on the 13Mp 5D test body from corner to corner. As expected the optimum apertures tend to be F5.6-F11, however F16 and F22 are still perfectly usable, with diffraction softening only seriously impacting quality at F32.|
|Chromatic Aberration||Chromatic aberration remains very low on full-frame, and indeed matches APS-C almost precisely. Again there's just a little green/magenta fringing at 70mm, and red/cyan at 200mm, with negligible CA in the middle of the range. You'll very rarely see any in real-world shots.|
|Falloff||We consider falloff to become perceptible when the corner illumination falls to more than 1 stop less than the centre. Maximum vignetting wide open with this lens is of the order of 2 stops at all focal lengths, with the pattern spreading progressively across the frame at longer focal lengths. The effect diminishes rapidly on stopping down, falling to about 1 stop at F5.6. Overall this is much what we would expect from this class of lens.|
|Distortion||Distortion is still low on full frame, at 1.2% barrel at 70mm, through neutral at 100mm, to 1% pincushion at telephoto. This will only very infrequently be visible in real-world photos.|
Full-frame compared to APS-C
Eagle-eyed viewers will no doubt have noticed that the MTF50 sharpness data at any particular focal length/aperture combination is distinctly higher on full-frame when compared to APS-C. This may at first sight appear unexpected, but in fact is an inevitable consequence of our presentation of the sharpness data in terms of line pairs per picture height (and thus independent of format size).
Quite simply, at any given focal length and aperture, the lens will have a fixed MTF50 profile when expressed in terms of line pairs per millimeter. In order to convert to lp/ph, we have to multiply by the sensor height (in mm); as the full-frame sensor is 1.6x larger, MTF50 should therefore be 1.6x higher.
In practice this is an oversimplification; our tests measure system MTF rather than purely lens MTF, and at higher frequencies the camera's anti-aliasing filter will have a significant effect in attenuating the measured MTF50. In addition, our testing procedure involves shooting a chart of fixed size, which therefore requires a closer shooting distance on full frame, and this will also have some influence on the MTF50 data.
Specific image quality issues
As always, our studio tests are backed up by taking hundreds of photographs with the lens across a range of subjects, and examining them in detail. This allows us to confirm our studio observations, and identify any other issues which don't show up in the tests.
The 70-200mm has a complex optical design, and the profusion of elements in its construction might be expected to result in some flare issues under unfavourable circumstances. In this respect it behaves somewhat similarly to its Nikon counterpart, generally handling backlit conditions perfectly competently, but occasionally running into serious issues with a strong light source either within, or just outside the frame.
Our two 'real-world' flare examples show this clearly (on our trusty but dusty EOS 5D); with the sun placed in the corner of the frame at 70mm, we see a long complex flare pattern, which shows up mainly as greatly reduced contrast with the aperture wide open, but progressively sharpens up on stopping down. With the sun placed just outside the frame at 200mm, we simply see an extreme loss of contrast through the image. To be fair this isn't a particularly unexpected performance for this type of lens, and whilst these issues are easy to demonstrate when explicitly looking for them, they're not exactly common shooting situations for most users.
|70mm F16, Canon EOS 5D||200mm F5.6, Canon EOS 5D|
Background Blur ('bokeh')
One genuinely desirable, but difficult to measure aspect of a lens's performance is the ability to deliver smoothly blurred out-of-focus regions when trying to isolate a subject from the background, generally when using a long focal length and large aperture. Here the 70-200mm F2.8 L IS does very well, producing smooth, 'creamy' bokeh which very rarely detracts from the subject.
|160mm F2.8, Canon EOS 5D||175mm F2.8, Canon EOS-1D mark III|
|50% crop||50% crop|
The issue of light falloff (or with this type of lens, more strictly vignetting) has become a topic of much heated debate on our forums, with some protagonists arguing that it will ultimately doom 35mm full-frame to an also-ran in the the battle of the formats, whilst those in the opposite camp opine that it simply doesn't matter outside of lab testing regimes. As usual the truth is somewhere in between; in many kinds of shots a little darkening at the edges simply doesn't matter, and can actually help draw attention to the subject (indeed edge-burning techniques have long been used in the conventional darkroom for this very purpose); however in some situations and applications it can genuinely detract from a photo. In reality, the answer in any specific case depends not just on the amount of vignetting/falloff exhibited by the lens, but also on a range of other factors including the shape of the falloff pattern, the evenness of the lighting, the subject, and the intended purpose of the image, so it's not surprising opinions vary.
|100mm F2.8 (2 stops falloff)||200mm F2.8 (1.7 stops falloff)|
As can be seen from these two samples, vignetting can be clearly visible in shots with evenly-illuminated areas such as the clear blue sky in the City skyline; in this case the question as to whether or not it is a 'problem' tends to be more one of aesthetic preference, rather than absolute right or wrong. In the second sample, variations of light and shade within the scene simply negate the effects of vignetting altogether. Now it's also fair to say that in certain specific types of photography falloff/vignetting is a genuine issue, but it does need to be kept in perspective.
Our studio tests indicate chromatic aberration to be extremely low with the 70-200mm F2.8 L IS USM, and this is indeed the case. In fact, in searching through hundreds of test shots with this lens across a broad range of conditions, we've struggled to find any shots exhibiting significant fringing at all. The samples below show the colour fringing you can expect to get at the extremes of the zoom range (there's simply none visible whatsoever in the mid range); note that these are shot with the rather demanding EOS-1Ds Mark III.
|70mm F8, Canon EOS-1Ds Mark III||200mm F11, Canon EOS-1Ds Mark III|
|100% crop, left side of frame||100% crop, top right corner|
Optical Image Stabilization
The 70-200mm features Canon's second generation vibration reduction system, which claims to allow hand-holding at shutter speeds three stops lower than usual before blur from camera shake becomes apparent. The mechanism is practically silent in use, with just a very quiet whirring noise when operational, and with just-audible clicks when it activates and deactivates from the IS group moving in and out of the 'at rest' position.
We've generally found the stabilization units in SLR lenses to be pretty effective in real-world use, and to quantify this, we subjected the 70-200mm to our studio image stabilization test at the wide and long ends of the range, using the EOS 5D as the test camera. With this combination we'd normally expect to be able to get good results handheld at 1/80 sec at 70mm, and 1/250 sec at telephoto without image stabilization. The subject distance for these tests was approximately 2m at 70mm, and 6m at 200mm.
We take 10 shots at each shutter speed and visually rate them for sharpness. Shots considered 'sharp' have no visible blur at the pixel level, and are therefore suitable for viewing or printing at the largest sizes, whereas files with 'mild blur' are only slightly soft, and perfectly usable for all but the most critical applications.
|70mm IS OFF||200mm IS OFF|
|70mm IS ON||200mm IS ON|
Not for the first time, we see Canon's in-lens IS system performing very well, and delivering an advantage close to the claimed three stops, especially at the telephoto end where it's needed most. One key difference here is not so much the yield of critically sharp shots at slow shutter speeds, as the hugely increased chances of getting usable shots with only mild blur, which would be completely impossible in the absence of stabilization.
At 70mm, we see a 70% chance of getting usable results at shutter speeds of 1/10 sec, and at 200mm, a 60% chance as low as 1/15sec. Even at speeds a stop slower than these, you'll get 'keepers' hand held if you're prepared to take multiple shots. This therefore greatly increases the lens's versatility in low-light conditions (or indeed any other situation where a slow shutter speed would be desirable).