Canon 35mm f/1.4 USM I vs USM II on the Canon 5D Mark IV and 5DSr - Resolving Detail Comparison
This is going to be a short write up...
As you recall, half a year ago, I did a review in which I compared the newer version Leica 35mm f/2 Summicron ASPH with the previous version. In that comparison, I was able to demonstrate that the newer version resolved details better at higher resolution.
Well... this time around, I have two version Canon 35mm f/1.4L - the USM I and the USM II - which I will compare with the Canon 5D Mark IV (at 30 megapixels), and the Canon 5DSr (at 50 megapixels).
Right off the bat, I'm assuming that the previous version 35mm f/1.4L USM I will resolve details similar to the current version USM II on the Canon 5D Mark IV. Furthermore, I'm also assuming that the current version 35mm f/1.4L USM II will resolve details better at higher resolution, on the Canon 5DSr. That would make the most sense. Otherwise, why would Canon release an updated version.
Canon 5D Mark IV + Canon 35mm f/1.4 USM I at f/1.4
Magnification. Canon 5D Mark IV + Canon 35mm f/1.4 USM I at f/1.4. For the first set, the lead eye is the right eye.
Canon 5D Mark IV + Canon 35mm f/1.4 USM II at f/1.4
Magnification. Canon 5D Mark IV + Canon 35mm f/1.4 USM II at f/1.4. Clearly the eye lashes here appear more refined than the USM I magnification sample.
Canon 5DSr + Canon 35mm f/1.4 USM I at f/1.4
Magnification. Canon 5DSr + Canon 35mm f/1.4 USM I at f/1.4
Canon 5DSr + Canon 35mm f/1.4 USM II at f/1.4
Magnification. Canon 5DSr + Canon 35mm f/1.4 USM II at f/1.4. I'm disappointed with this sample. Something must have went wrong, since the difference isn't immediately perceptible.
Of course, you know what they say when you assume. Needless to say, I was not exactly happy with the results. The USM II version did much better than the USM I version on the 5D Mark IV - which I didn't expect. Furthermore, the USM II version didn't appear to resolve details noticeably better than the USM I version on the 5DSr? That was not suppose to happen.
Maybe it's because I shot everything wide open at f/1.4? Maybe my gear wasn't accurate. Maybe my shooting fundamentals were just not very good? All sort of second guessing entered my head. But it didn't make sense. My hands were steady. I used mirror lock up. My elbows were prone on the table. I was even seated. Plus I relied completely on the auto focus. It just didn't make sense.
All that work for nothing I thought! It was then, while examining my image files under post, that I decided to do a follow-up test. But this time, I was going to shoot closer and at f/4... and only on the lead eye.
So much for a short a write-up. This is what happened on the follow-up day.
Canon 5D Mark IV + Canon 35mm f/1.4 USM I at f/4
Magnification. Canon 5D Mark IV + Canon 35mm f/1.4 USM I at f/4. The tips of the eye lashes are clearly feathered - evidence of reduced detail.
Canon 5D Mark IV + Canon 35mm f/1.4 USM II at f/4
Magnification. Canon 5D Mark IV + Canon 35mm f/1.4 USM II at f/4. By comparison, the eye lashes on this USM II version has greater clarity and refinement.
Canon 5DSr + Canon 35mm f/1.4 USM I at f/4
Magnification. Canon 5DSr + Canon 35mm f/1.4 USM I at f/4. Strangely, the eye lashes look pretty good from this USM I version at 50 megapixels.
Canon 5DSr + Canon 35mm f/1.4 USM II at f/4
Canon 5DSr + Canon 35mm f/1.4 USM II at f/4 - Magnification. Clearly, the details of the eye lashes on the 5DSr from the USM II version is the most refined. Plus, Anna's baby blues just pops out in this combination. OMG, that is amazing!
Upon examination, the Canon 35mm f/1.4L USM II resolved details the best, which can be seen on the Canon 5DSr. You can tell by how much detail is retained in Anna's eye lashes. In fact, there was even a noticeable improvement in detail of the USM II version on the Canon 5D Mark IV.
So, there you have it, the Canon 35mm f/1.4L USM II resolves details better at higher resolution. You can expect Canon to update their entire lens line up in the near future. You know it's going to happen, since it's an untapped source of revenue for Canon.
All the images in this writeup have only had their color balance optimized in Lightroom. Contrast, clarity, vibrance and saturation were uneffected. Images in the first set have been leveled slightly, and therefore cropped. Obviously all the magnification samples were the result of cropping.
Thanks again to Anna. You're the best!
