Originally posted by bdery Tossing one number in the discussion isn't very helpful.
Au contraire: It negates the claim that the lens cannot deliver corner sharpness.
And ironically, that number is taken from the negatively worded Optyczne/Lenstip review that @normhead linked.
Originally posted by bdery You can of course find one setting where the lens is at its best. The same is true of any lens.
At its best for one particular parameter, at least.
But the issue is, how good is that best?
The A 20/2.8, for example, has soft corners on APS-C at all apertures,
so I do not find that its best in that regard is good enough for critical applications.
Originally posted by bdery Trust me, I'm an optical designer
I am aware of that, which is why I quoted you a specific number,
which might not have meant much to casual users,
rather than saying, for example, that it delivers more than 90%
of the optimal corner sharpness on APS-C available from the FA 31.
Originally posted by bdery But I'm talking about sample images.
Which ones? The ones from the very start of production?
Originally posted by bdery I could not see myself replacing my 21 and 40 with the 20-40, even though I would really, really like to do so.
No need to get defensive, unless you designed the lens, of course.
It is not a question of "getting defensive," but of setting the record straight.
It would be a shame if potential users of this lens were misled by the kind of ill-informed comments that have appeared,
or thought that the early samples are representative of what you would get if you were to purchase the lens now.
I was very skeptical of this lens when I first received it,
precisely because of the experiences of early users like Adam or Sandy Hancock,
so I was very pleasantly surprised when I found it to be better
than my A20/2.8, K24/2.8, and DA 40 XS.
And since it is a zoom, after all, I am not disappointed that it doesn't outperform my DA 35 Ltd.