Review Canon 35mm f/2 IS (C APS-C)
The Canon EF 35mm f/2 USM lens is Canon's first lens with a fixed 35 mm focal length and a built in Image Stabilizer. This lens is designed for full frame camera's and is available since December 2012. Since 1990, a Canon 35mm f/2 lens without IS is available, which is cheaper and weighs less.
Canon EF 35 mm f/2 IS USM review @ Canon 650D
Canon Mark II lens cap
Construction and autofocus
|The lens is very solidly built and the mount is metal. The lens has a window in which you can see the distance where the lens is in focus. There is also a depth of field scale engraved on the lens. Which can, however, only be used in combination with a full-frame camera. The Canon 650D test camera, with its smaller APS-C sensor, has an even greater depth than shown on the depth of field scale.
The autofocus is quiet and fast, thanks to a ring-type USM (Ultra Sonic Motor) and - according to Canon - a powerful CPU. The front lens does not rotate during focusing. Manual focus is always possible by simply turning the focus ring. You don't have to switch the lens to MF first. Both AF and IS are very silent. Even in low light, the AF is fast.
No lens correction
|Normally, we try to apply as many in-camera corrections as possible to the jpg files stored in the camera. The test of the Canon 35 mm 2 IS is performed on a Canon 650D. The camera indicates that for this lens there are no correction data for vignetting and chromatic aberration available. Our Imatest measurement results for jpg files, where the lens aberration corrections (vignetting and chromatic aberration) were turned on, were consistent with the results of RAW files without lens aberration corrections.
The Canon 35 mm f/2 IS lens is two stops faster than a kit or zoom lenses, with f/4 at a 35 mm focal length. This means that with the Canon 35mm f/2 IS - under otherwise equal circumstances - you can use a 4x faster shutter speed compared to these zoom lenses. And if you use the -virtually silent - image stabilization, you will win another 3 stops. Even in very low light, you can now shoot images without the use of a tripod. The image to the right was shot with a shutter speed of 1/4 second.
Move your mouse over the image on the right, for a 100% crop of the green frame.
Here you see our test results for the image stabilization: when you take a look at the sharpness of an image taken with a shutter speed of 1/6 second, it seems that Canon's 4-stop claim for the image stabilization, is not simply a slogan, but fully justified. Very good!
Canon EF 35 mm f/2 IS USM Resolution
|This Canon lens offers high sharpness from center to edge, at every aperture. The sharpness at f/2 is as high as the sharpness some kit lenses realize after stopped down 1 or 2 stops (i.e. f/5.6 - f / 8). There is not always enough light to do this, without the without risk of blurred images. But the sharpness of the Canon 35mm f/2 IS is not unique. In terms of sharpness, this lens has several competitors, such as the Canon 40mm STM, the Sigma 35mm f/1.4 and the Sigma 17-70 mm f/2.8-4. We haven't tested the old version of the Canon 35mm f/2, without built-in image stabilization, yet. But that lens probably will also be on the list of potential competitors.
In direct comparison of 100% crops of images taken at f/2.8 and f/8, it is just visible that the image shot at f/8 is a bit sharper (look at the dark feathers). It is a difference which I never would worry about. The most important parameters for choosing an aperture, will be the desired depth of field and the available amount of light.
Click on the right picture.
|The Canon 35mm f/2 IS is designed for use on a camera with a full frame sensor. If you use this lens on a camera with a smaller APS-C sensor, you use only the center of the image and the vignetting is substantially lower. Only at f/2 there is a remnant of visible vignetting, but at all other apertures vignetting is completely absent.
Distortion Canon EF 35 mm f/2 IS USM
|There Canon 35mm f/2 IS lens shows 1% barrel distortion on a camera with an APS-C sensor. On a camera with a full frame sensor (Canon 5D MK3) we measured 1.5%. This is a typical value for this type of lenses, as we saw in our recent example, the Sigma 35mm f/1.4 review. In most cases, this level of distortion is not noticeable. In the most critical situations, you may choose to post the distortion correction software.
A relatively high bright give some extra leeway to play with depth. Also, the bokeh is not bad for a wide angle lens on a camera with an APS-C sensor. Yet this lens loses the terms of bokeh light stronger Sigma 35mm f/1.4 and possibly the Canon 35mm f/1.4, which we have not yet tested.
Click on the image opposite for a comparison of the bokeh of this lens at full aperture on a camera with a full frame sensor (Canon 5D MK3) and a camera with an APS-C sensor (Canon 650D).
Canon 35mm f/2 IS @ f/2, 1/8 sec
|Every single element has a Super Spectra coating to reduce glare and reflection. And the Canon 35mm f/2 IS is indeed very insensitive towards flare. If there is no light source shining directly into the camera, then this lens is almost unaffected by flare. Even when the sun shines directly into the picture, there is virtually no ghosting and glare is limited to a small area around this bright light source.
So perhaps there's no real need to use a lens hood. Yet, for a lens in this price range, you expect it to came with a lens hood. That is not the case.
The Canon EF 35mm f/2 IS USM has a molded aspherical lens element at the rear of the lens to correct for chromatic aberrations of the entire optical system. Well done, Canon. Both the RAW and JPG files show in practice no visible chromatic aberration.
Conclusion Canon 35mm f/2 IS review
Canon 35mm f/2 IS review RAW & JPG score: Normally this table shows the performance of this lens when you save the files in the camera as jpg, with all available in-camera lens corrections turned on. The score will give you for this lens/camera combination: "What you see is what you get". But our Canon 650D test camera had no correction data available for this lens. Therefore the results for jpg files and uncorrected RAW files are almost equal.