llcFM Offline Image Upload: Off
|
p.1 #16 · DigLLoyd's Subscription Site | |
Thank you everyone for the compliments. I am going to keep working hard to keep providing good value.
Regarding sample variation, I agree that one cannot come to a statistical conclusion with only a few samples.
Grouping a fairly large number of new lenses of each brand (Canon, Nikon, Zeiss), I do have a pretty strong opinion on the quality variation, which is to say that Canon and Nikon have had frequent problems for me, but Zeiss is significantly better. I've now seen 20 or so Zeiss ZF lenses, which is a reasonably large group, even if they aren't all the same focal length, and I've seen more than that of Canon/Nikon. The occasional variation I've seen with Zeiss has required very precise testing, whereas it has been fairly obvious in the Canon/Nikon cases. Of course, the ZF lenses are not zooms, and thus less complex, and therefore easier to build consistently.
My Brand-new Blur article covers a few of the bad samples I've run into. I’ve had trouble with quite a few others, including the Nikon 17-35/2.8 (so blurry on one side it could be seen through the viewfinder), and the Canon 24-70/2.8L and 16-35/2.8L.
I don't think Canon is any worse than Nikon, but I've had too many problems to trust any brand. When I get a new lens, I test it with field shots for a quick check (certain revealing subjects), then I do something more rigorous. It's a tedious and annoying task, but better than the alternative of shooting a bunch of frames that are soft on one side.
Both Canon and Nikon are selling full-frame digital cameras that show any optical problems clearly. I suspect that quality (less sample variation) might be getting better, but then again it might not. It's just way too costly to make sure 100,000 lenses are within 0.1% variance.
|