Canon EF-S 10-22 mm f/3.5-4.5 USM
4. Image resolution
Is such a high result in this category a fact or is it an effect of some systematic error in our tests? We can answer this question by comparing the behavior of a couple of lenses in diffraction limit, that is, for example with an f/16 aperture, where the off axis aberrations are minimal and the only influence on the resolution would be by diffraction, the detector quality and the systematic errors. Thus for the 17 and 22 mm focal lengths the average MTF50 value is 31.6+/- 0.5 lpmm. The error given here is the 1-sigma gaussian error coming from the average procedure from over dozen of measurements.
In the case of 1.8/85 and 1.4/50, Canon's average MTF50 value for f/16 was exactly 29.7 lpmm with an error, respectively, of 0.8 and 0.3 lpmm. We can see that the MTF values for the EF-Sa 10-22 mm are 6% higher. On the other hand, however, according to the statistical inference both tests coincide in three (or actually two) sigma values, which does not mean in any sense that the EF-S 10-22 mm tests we conducted were erroneous. However, we cannot completely rule out the claim that the real resolution values are a few degrees smaller than the ones we provided in our tests. Nevertheless, it does not change our opinion that the Canon 10-12 mm has a very good picture resolution.
Please Support UsIf you enjoy our reviews and articles, and you want us to continue our work please, support our website by donating through PayPal. The funds are going to be used for paying our editorial team, renting servers, and equipping our testing studio; only that way we will be able to continue providing you interesting content for free. |
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Another argument supporting the precision of our test methods is the fact that the Canon does not perform so well on at the edge of the frame. Therefore, in the middle of the focal lengths range the performance is very good, where the lens has no problem exceeding the 30 lpmm level. The Canon's performance on the focal lengths range endings is worse, but still acceptable.
To be consistent, we checked the lens's performance in diffraction limit for an f/16 diaphragm. The Canon 10-22 mm reaches values of 25.8-26.7 lpmm with an average error of 1 lpmm. Prime Canons reach 28 lpmm with an error of less that 1 lpmm in this category. This time the test results coincide with the standard gaussian deviation. This is a strong argument supporting the fact that if our resolution tests for wide lenses carry some kinds of systematic errors, than it is a very slight error, which has no influence whatsoever on the opinion we have established that the Canon EF-S 10-22 mm has great picture resolution.
