The trouble is, with a sample size of one, a test proves almost nothing, unless the manufacturer's QA processes have been shown to be good enough for variation to be very small.

I don't think it would be in Pentax's interests to cherry-pick either the very best, or the very worst - if they picked the best, the test would be great, but then everyone would be disappointed, and their reputation (as well as the tester's) would take a pasting as soon as more people started doing their own tests.

Whereas if they picked the worst, the test results would be rubbish, and no one would buy the lens.

So the best thing would be to pick one that's roughly in the middle of the range they consider acceptable for release. Then the test will be meaningful, and the real-world results will bear that out, with some better and some worse - hopefully only a little bit better and worse, if the QA is any good.

What you wouldn't then get, is what we have, which is a lot of anecdotal so-called evidence (the DA* 16-50 comes to mind) that calls the QA process into disrepute, and confidence tends to be lost.

Quite so, and this is the depressing thing about it all. You'd think the magazine / lens manufacturer would ensure that the one reviewed would be a good one, unless they've got something to hide.

Any idea how it compares with the 1.7 at 1.7 and smaller? If there's not much in it then I'll definitely cross the manual focus 1.4 off my list of "maybe sometime".

Soft?

If Galoot's example is typical (bearing in mind that it's about a quarter of the size of the original), then it's not soft, it's c**p!!!! I'm still hopeful that it's technique at fault. No discredit, just that in general I'd rather it's my being rubbish than the kit's, if you know what I mean.

There's something funny going on, I'm sure. Haven't we seen examples that were much sharper than this at F1.4?

I may be making all this up, but I have a memory that it was Matt's FA 50 1.4 - isn't it supposed to be optically the same as the A-50?

I was sure you did.

I've been put right off getting the 16-50 by the reports elsewhere of the variability in its quality.

I can hear George tutting even as I type, but sadly it's not his £500 I'm thinking of spending...

This is a question I've been trying, unsuccessfully, to answer for myself as I weigh up whether to get a DA* 16-50, since unless it's good at F2.8, I don't want to pay £500 for it.

I had a quick look at the online DOF calculator which suggests that the DOF when focused, say at 50 feet, is such that it's acceptably sharp only from about 40 to about 60 feet at F1.4.

I can't really tell how far away your subjects are, but that's not much DOF at that distance. If it's focused at 30 feet, the DOF is only +/- 3 feet!

That said, your first picture doesn't look all that sharp anywhere, so I have no good answers really. At 1/320s, there shouldn't be any camera shake, but who knows. I assume you took several at that sort of exposure, and they were all much the same.

I have to say, I have nowhere near enough confidence in how to go about assessing these things, to go out and buy an expensive lens.

But if I was going to try and do tests like this, I'd be using properly repeatable conditions, which this is clearly not.

Nor with RAW. When it's gone, it's gone.

I alternate between two batteries in exactly the way John describes, and have also found that the one in waiting retains its charge very well, showing no noticeable discharge over as much as a month.

One thing to watch though, is the level indicator on the K10D, which is almost useless. Presumably because the battery is so good at supplying its power right up to exhaustion, you get practically no warning from the level meter when it's about to expire.

On mine it goes from full to empty over 1 or occasionally 2 shots, so make sure you carry the charged spare with you.

Quite. Presumably the number of seals relates to how many things need sealing. It would be a bit daft to seal only some of the bits that can let water in.

Bless.