 |
Grevillea buxifolia hand held RF 100mm f2.8 L Macro |
Released in 2021, The Canon RF 100mm f2.8 L Macro is one of the best lenses I have ever owned. It renders superbly sharp images across the frame and into the corners right from f2.8, with barely any evidence of distortion or aberrations. In addition it features super fast, accurate autofocus making it suitable for sports, action and even birds in flight.
I have read or viewed 13 reviews of this lens in addition to my own reviews on this blog. Most express fulsome praise for the lens with few if any downsides.
But two reviewers have reported focus shift.
To explain: Canon lenses focus with the aperture wide open for optimal speed and accuracy then close down to the selected aperture for the exposure. This means the designers need to make sure the plane of best sharpness does not shift as the aperture is closed down.
Two of the reviewers to date have reported some kind of focus shift.
The first is Bryan Carnathan of The Digital Picture (website) who reported focus shift at high magnification levels. I note however that at least as reported, his testing procedure appears to have been more naturalistic than systematic.
The second is Dustin Abbott of his eponymous Youtube channel. Mr Abbott reported some unexpected variation in sharpness at the center of a test chart at normal distance and attributed this to focus shift. However he did not report systematically testing for focus shift using a bespoke method like the one which I use as described below.
I tested for distribution of the zone of sharpest focus using my usual method which you can read about here.
I wish more lens testers would use this method as it can quickly reveal much useful information about what happens to the three dimensional distribution of sharpness as the aperture is closed down. There are many important characteristics of a lens which cannot be revealed by photographing flat test charts.
I find the RF 100mm f2.8 L Macro gives an exemplary performance on this test. Sharpness is distributed evenly across the frame with no detectable shift as the aperture is altered.
 |
Photo of grass with Adobe Find edges filter. F2.8 Focus is on the top edge of the stick I find no focus shift as the aperture is reduced. Notice sharpness is distributed evenly across the frame to the edges. |
To test for focus shift at macro distances, I made up the little test rig you can see in the photo.
I align the lens axis perpendicular to the target face of the test subject and level both test rig and camera from side to side. I photograph the target at f2.8, f4 …etc up to f32.
I did not detect any evidence of focus shift on this test.
 |
Focus shift test jig for close-up distances |
Summary and Conclusion
I set up test protocols specifically designed to detect the presence of focus shift both at normal distances and macro (close-up) distances.
I found no evidence of focus shift.
The characteristic movement of the zone of best sharpness of a lens with aperture change (and focus distance change and with zooms focal length change) is baked into the design of a lens as are things like distortion and peripheral shading.
I think it unlikely therefore that sample variation could account for significant differences in results from independent reviewers. Any difference seen is more likely to be due to variations in test procedure.