Studio Tests (Full frame)
The Canon EF 40mm f/2.8 STM again gives very good results on full frame, producing remarkably consistent results from F2.8 through to F11. Central sharpness is very impressive, and although the corners never quite catch up, they're still more than respectable. Distortion and chromatic aberration are both very low, and vignetting only really visible at F2.8. Once again, the 40mm pretty closely matches the results from the EF 50mm f/1.8 II or the EF 35mm f/2.0, and even the much more expensive EF 50mm f/1.4 USM.
|Sharpness||Sharpness is very high, and impressively even across the frame even at F2.8. It stays remarkably consistent at all apertures to F11, beyond which diffraction starts to soften the image. But even F22 is still entirely usable if extreme depth of field is required.|
|Chromatic Aberration||Chromatic aberration is very low - indeed recorrection towards the corners means that if anything it's less visible on full frame compared to APS-C. There's really nothing to worry about here.|
|Vignetting||The 40mm shows about 1.7 stops vignetting at F2.8, which is more than you'll get from faster lenses stopped down to the same aperture. However it has a relatively broad falloff profile, which means it's not as visually objectionable as lenses which show abrupt vignetting in the corners of the frame.|
|Distortion||Distortion is still very low on full frame. There's a tiny amount of barrel distortion (0.7%), but this will rarely be visible in real-world use.|
As you'd expect, it's a similar story on full frame to what we saw on APS-C when shooting at the lens's minimum focus distance. The centre of the image is pretty sharp at F2.8, but the corners are distinctly soft and only fully sharpen up at F16 in this flat-field chart test, suggesting some curvature of field. There's slight, but visible barrel distortion, and a little fringing from lateral chromatic aberration.
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 EF 40mm f/2.8 STM generally performs very well, producing high quality images in a wide range of situations.
The 40mm's compact design and exposed front element means there's very little shielding of the optics from peripheral light sources. In general it's reasonably well-behaved with the sun in the frame, but can be somewhat prone to flare when shooting into the light, and in bright conditions it's advisable to use a hood (as usual). Disappointingly Canon doesn't provide one with the lens, but instead offers the optional slimline ES-52 hood for around $29 / £20 / €20 (which has of course been industriously cloned by Chinese companies such as JJC). Alternatively you can use a generic 52mm screw-in hood.
The first example here is a deliberate torture-test, and pretty much the worst-case scenario (the sun is just slightly outside the frame). Generally in such situations you can see what's happening in the viewfinder, and either adjust your composition, or provide a bit of extra shading to the front element to reduce the effect. The second example illustrates what happens which you shoot directly into the sun on APS-C at a more usual working aperture - here contrast has held up very well.
|Canon EOS 6D, F22, sun just outside frame||Canon EOS 650D, F8|
The studio tests show that the 40mm exhibits very little colour fringing from lateral chromatic aberration on full frame, and this is borne out in real world use. You have to look really, really hard at your image files to see any, and you've have to make enormous prints for it to be a problem. Canon's most recent SLRs will correct it in their JPEG processing too.
For the record, here's an example that illustrates the 40mm's characteristics, taking 100% crops from RAW conversions with and without CA correction applied. There's a tiny amount of fringing visible in the uncorrected example - really nothing you'd normally worry about - which can be removed almost completely if necessary.
Canon EOS 6D, F8, 1/100sec ISO 200
|RAW + ACR, no correction||RAW + ACR, CA correction applied|
|100% crop, top right corner||100% crop, fringing removed|
The tests show that the 40mm f/2.8 exhibits about 1.7 stops vignetting at maximum aperture - this isn't unusual from a full frame lens shot wide open, but it does mean you get more vignetting compared to shooting a faster lens stopped down to F2.8. But the fairly gentle falloff pattern means it's not especially intrusive - in many real-world images it will disappear into the natural variations of brightness across the frame.
All of Canon's full frame SLRs made in the past 5 years or so (back as far as the EOS 5D Mark II) can correct vignetting in JPEG processing if necessary, and the rollover below compares an image shot at F2.8 using the Canon EOS 6D with a version that's had the vignetting corrected using the in-camera raw converter. There's no inherently right or wrong answer here - which you prefer is very much a personal choice.
|Canon EOS 6D, 40mm F2.8, uncorrected||40mm F2.8, corrected|
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/or a large aperture. The 40mm F2.8 is perhaps not the ideal choice for blurring away backgrounds - if this is what you're after, a longer or faster lens will work better - but its rendition of out-of-focus backgrounds is generally very attractive, with a smooth, natural look.
This is in marked contrast to the pentagonal highlights produced by the EF 50mm f/1.8 II at similar apertures (see here for examples).
|EOS 6D, F2.8||Background crop, upper left|
|EOS 100D, F2.8||Background crop, upper left|