The 400L samples had obvious issues, I think he got called on it.
The only "issue" was that he forgot he had a filter on the 400L and didn't on the Tamron. I don't think he got "called" on it. Someone pointed out the mistake and he corrected it. It's debatable how much an impact the filter would have, but give him credit for recognizing the issue and being straightforward about it.
The Tamron was a loaner from Tamron, and they only loan out the really good ones, no matter what they tell you. They'd be pretty stupid if they didn't select the best of the best for loaners. Then, they put them on a shelf and seem to be grabbing one at random, when they are actually all hand picked lenses.
These statements are common on the internet, but I wonder if anyone has any actual, personal proof or experience with this. Or, if it's just something that gets repeated and since it sounds logical, people believe it.
Trying to think it through logically. Let's say the lens comes from the in-country distributor. Do they really pull out 10 or 20 lenses from their stock and run a battery of tests on each one before loaning one out to a reviewer? Do they even have the necessary equipment to run these tests? And, do they have the personnel with the time and expertise to pre-test these lenses? Does anybody know this, or is this just conjecture?
And, while we're on the topic. It seems like Canon and Nikon would be more likely to do this than a third-party manufacturer if only because they'd be more likely to have the resources available for this kind of manipulation.
I would agree that testing a pre-production copy would be problematic, because it's a product that will never get into the consumer's hands and many times these pre-production copies are small runs that are indeed assembled and tested under more scrutiny than a production run lens. But if a distributor has, say 2,000 boxed production-run copies of a lens, what evidence is there that they are pre-testing these lenses. And, if you think they will go to that extent, wouldn't they also pre-test any lens they sell to a well-known review site?
Unless the reviewers hire a third-party to anonymously purchase a lens from a retailer, the exact same cherry-picking can occur. Is that the procedure followed by review sites? Somehow, I doubt it.
And, while we are thinking about this, wouldn't it be in the best interests of the in-country distributor to hand over a production-run, untested copy? The last thing they want is to sell 10,000 copies of a lens and have 8,000 returned because someone fudged the reviews. If I'm a local distributor and my company produces a product that turns out to be a turkey, I want to have that known as quickly as possible so I have more leverage to force the parent company to make it right. Cherry-picking a review sample just is not in my best interests.
Now, I would say it's a good idea to wait until several reviewers have weighed in and frankly, I would want to wait until someone like Roger Cicala has a chance to run several dozen or several hundred copies through his system so they are thoroughly field-tested and any problems can be identified before buying. But I don't like blanket statements without any evidence to back them up.
I'm a little surprised at how many people have already made up their minds about this lens before it has really hit the market. I strongly suspect that the reviews won't really matter. Those who are inclined to hate it will find reasons to do so, those who are inclined to love it will find reasons to do so.