The common thinking seems to be that there's a significant variation between individual identical lenses, there being good and bad copies. It would be very interesting to see some systematic testing of this. Like for example testing 5 identical primes lenses of one brand at a time. I guess it's not too easy for Photozone to do since you have to rely on lenses lended by individual people and the manufacturers may not be that eager to let you test lenses like that if they suspect they might be significant variation.
Do you know if any site or magazine has done this kind of testing?
[quote name='emr' timestamp='1307985905' post='9180']
The common thinking seems to be that there's a significant variation between individual identical lenses, there being good and bad copies. It would be very interesting to see some systematic testing of this. Like for example testing 5 identical primes lenses of one brand at a time. I guess it's not too easy for Photozone to do since you have to rely on lenses lended by individual people and the manufacturers may not be that eager to let you test lenses like that if they suspect they might be significant variation.
Do you know if any site or magazine has done this kind of testing?
[/quote]
a few tests have been done. lenstip.com tested several (was it 30 or 50) mm f1.4 lenses, but on different mounts.
And I think dpreview ? had also tests with several copies of 1 or 2 lenses, I can't find it though.
[quote name='emr' timestamp='1307985905' post='9180']
The common thinking seems to be that there's a significant variation between individual identical lenses, there being good and bad copies. It would be very interesting to see some systematic testing of this. Like for example testing 5 identical primes lenses of one brand at a time. I guess it's not too easy for Photozone to do since you have to rely on lenses lended by individual people and the manufacturers may not be that eager to let you test lenses like that if they suspect they might be significant variation.
Do you know if any site or magazine has done this kind of testing?
[/quote]
Well, we do test some samples if necessary. However, what is the point ? It is not sufficient to test 2, 3 or even 5 samples to get an idea about the statistical variations.
[quote name='photonius' timestamp='1307991555' post='9183']
a few tests have been done. lenstip.com tested several (was it 30 or 50) mm f1.4 lenses, but on different mounts.
And I think dpreview ? had also tests with several copies of 1 or 2 lenses, I can't find it though.
[/quote]
Here is a good read, and it links to the tests (was slrgear actually, not dpreview)
http://www.lensrentals.com/blog/2010/03/this-lens-is-soft-and-other-facts
' />