After testing a set of brand new 28mm lenses for my Nikon 28mm f/1.8G Review a couple of weeks ago, I was rather disappointed by the overall performance of the lens. Both samples that I tested exhibited visible focus shift and field curvature issues, which impacted performance in a “wavy” pattern. This weekend, I decided to give another Nikon 28mm f/1.8G a try and see if it has the same optical issues (borrowed from our team member Bob Vishneski).
To my surprise, the third lens sample performed much better in comparison to the first two. Here is the original chart that I published in my review:
And here is an updated chart from a “good” sample:
While the mid-frame stayed about the same, the significant improvement in the center and the corners is clearly visible here. Unfortunately, the third sample also showed some field curvature, which now I am 100% confident is a lens design issue. In fact, if you look at the above chart, you will notice that the corners on the lens are better than the mid-frame, which is an indication of a donut-shaped field curvature (sharp in the center, less sharp in the mid-frame and sharp again in the corner).
What does this all mean? Lenses, just like camera bodies, can vary in performance. Most lenses are excellent out of the box, but some can be weak. Aside from optical design issues, lenses can have different problems – from optical lens element shifting and spacing issues to mechanical and electrical problems. The good news is, there are definitely a lot more good lenses out there than bad, especially when it comes to Nikkor glass. During the last 4 years of lens testing, I rarely had bad lens samples (and I have tried 100+ Nikkor lenses by now). Maybe 3-4 out of 100 were faulty. But as it turned out in this case, I am not that lucky sometimes, having tried two 28mm f/1.8G samples that both had performance issues. Sometimes bad lenses are not even related to quality assurance mishaps – packages get dropped all the time and that alone can damage lenses. Don’t forget, we are dealing with high-precision instruments here; even a slight variance can impact the overall performance.
I am not saying all this to scare you, but to make you aware that sometimes performance data from lens reviews can be rather useless. How does one know that the lens being reviewed is a good sample? When reviewing lenses, I always prioritize my real-life experience with a lens over lab tests. I spend about a month shooting with a lens and leave all lab tests towards the end. Shooting test charts and other targets is very boring, but lab tests are necessary – that’s how you know how one lens compares to another and that’s how you know if a lens has an optical problem. Without these lab tests, I would not have been able to tell you that the newer and cheaper Nikon 50mm f/1.8G is better than the more expensive Nikon 50mm f/1.4G, for example. However, after wasting my entire weekend testing a bunch of lenses (which is a very time-consuming process, due to the complexity of the process), I have been questioning the need to do this kind of detailed analysis. Now I have to update the Nikon 28mm f/1.8G Review with all the new data, change my lens rating and recommendations. All this while I currently have 5 lenses to review, with 3 more on their way! Ouch…
More to come!