Quote: Interesting results - the D FA 15-30 and 24-70 seems to be about perfect, while the D FA* 70-200 is quite good but with poor corners wide open. The D FA* test appears to disagree with ephotozine's:
Pentax HD PENTAX-D FA* 70-200mm f/2.8 ED DC AW Review
The testers can't agree with each other, yet we are supposed to look at their work as definitive... just saying.
Quote: and my experience, as well - no way is it inferior to the 24-70 - I think they need another colour for 'stunning' for the best resolution of the lens
Maybe they should have borrowed your lenses for the test.
I'm suspicious of anyone using a sample size of less than 10 lenses...
I'm dismissive of anyone testing less than 5.
What I want to see in a lens test, is the result from testing with a grouping. What is the best I might get? what is the worst I might get? Is there a cluster of lenses in the middle of the curve and if so what are the averages across that cluster? Because that would be what I would be most likely to get.) That would be lens testing. This nonsense, not so much.
This type of testing is often available form lensrentals.com for the lenses they rent out.
And even they don't go ito the important stuff. From what lenses do people actually prefer the images and how strong are those preferences? Lens performance is subjective, a lens could have all the best specs ever, and still not be liked by a lot of people. Until these tests are couple with how real human beings respond to the images, they mean nothing.
Anyone testing one lens, has information on that lens. While that may be more than most of us know, it still is not enough to draw a conclusion about the group of lenses with that designation taken as a whole.
The thing is, I never have a problem looking at these tests and saying my copy is better or worse than that. Sample variation is known it's documented, and once lenses get out of the lab and experience some real life bumps and rinds, like the FA*80-200 they tested, it only gets worse.
I know folks like DxO and these magazines pump this rogue testing for all it's worth, but, trust it at your peril. It's not real science. It's science applied without knowledge of statistics or probability in single subject trials. Imagine if you tried to get a drug approved with trials like that .
Buy your lens for the price, quality, focal lengths and aperture that you need, and make it work. The thing I find most irritating, these guys are researchers, they know the limitations of their tests. yet , no fine print, no disclaimers, throw it out here and make like you're some kind of authority. I guess that's what makes them money.