Quote:... the Sony E 35mm f/1.8 OSS that I purchased for testing is ... yeaaaahhh ... heavily decentered. :wacko:
I'll have to do a relaxing test next - the Oly 75/1.8. I should stay away from this image stabilized crap for a while. Which reminds me that I should do some Zeiss ZE testing on Canon full format.
Why don't you just test the lens as it is? If you're screening for a better lens the rest of us buying the stuff might get "fooled"?
Quote:Why don't you just test the lens as it is? If you're screening for a better lens the rest of us buying the stuff might get "fooled"?
A review should give an indication about the performance of a non-faulty product. Because that's what you expect when you buy such a lens.
Testing a decentered lens (and publishing the results) would be like testing the performance of a car with flat tires.
-- Markus
Quote:A review should give an indication about the performance of a non-faulty product. Because that's what you expect when you buy such a lens.
Testing a decentered lens (and publishing the results) would be like testing the performance of a car with flat tires.
-- Markus
Two of the 10-18/1.4 and now one 35/1.8 OSS are decentered as first buys. What are the odds that these are defective and not bad by design? 1%? less?
Quote:Two of the 10-18/1.4 and now one 35/1.8 OSS are decentered as first buys. What are the odds that these are defective and not bad by design? 1%? less?
The design is Ok, but the manufacturing quality is dreadful.
You are correct in the sense that the decentering rate of Sony E OSS lenses seems to be worse than during the dark ages of Sigma.
As of now I cannot recommend the system.
Quote:A review should give an indication about the performance of a non-faulty product. Because that's what you expect when you buy such a lens.
Testing a decentered lens (and publishing the results) would be like testing the performance of a car with flat tires.
-- Markus
But if that is the level of Quality the manufacturer deems acceptable to be released "in the wild " it is more than fair to release the results "into the wild" as well.
Klaus has done so with the Pentax 50-200. Perhaps Sony also deserves this treatment. In particular if the trading (non-)culture in the continent with the kangaroos makes it difficult to select good samples.
Slrgear publishes results from lenses that are subpar - they normally comment on it. So you wouldn't be the first ...
At least traditionally - the price differential between an expensive lens brand (e.g. Zeiss) and a not so expensive brand was affected by the quality control - how many lenses of a batch were classed as rejects and distroyed instead if boxed and sold.
Perhaps a better way would be to test a number of each and publish the weakest instead of the best. Can be more work than testing through 3 million rejects before finding a good sample
Enough of brainstorming now.
To be fair with Sony - the Pentax 50-200 case was AFTER servicing by Pentax.
If Sony returns the 35 OSS with the same statement (within specs) I will surely publish the findings.
Quote:A review should give an indication about the performance of a non-faulty product. Because that's what you expect when you buy such a lens.
Testing a decentered lens (and publishing the results) would be like testing the performance of a car with flat tires.
-- Markus
I second that! The test should show the potential of the lens:
that's what a good copy can deliver. And bad copies should be mentioned in the text, as a caveat.
(And I believe this is how photozone usually does it.)
The prototypes were much sharper than the first production batch. Sony has a problem with first production batches, this is not the first time that the quality increases later. Lensrental has maybe data about it.