Indeed. The lines are a lot straighter at 12mm, even at the edges. I'll post some proof when I get a chance to set it up properly.

I need to find a brick wall to compare the sharpness with Simon's one; that won't be quite as easy.

How close is very close? I was about 12 feet from the house in my picture above.

Are they the same ones I saw after you put the links in, or now re-shot as I suggested above?

Couldn't agree more. The lack of sufficient information available to allow people to make informed decisions is scandalous, really. I suppose the manufacturers don't want to give up their secrets if they're not all positive.

The Sigma shows virtually no PF at all, even in the sort of conditions that encourage it - like tree branches as you say.

As you say, it's horses for courses.

When I get a chance I'll throw together a little comparison of its distortion at 10mm and 12mm, since to be fair, the Pentax only goes down to 12.

10mm is a lot wider than 12mm, it has to be said. For the sake of a bit of (very easy) correction sometimes, the extra width, lack of PF, and the hugely smaller price, I opted for the Sigma. So far, touch wood, I haven't been disappointed, although before I realised how to make sure the focusing was happening, I was a little concerned.

What's it like for purple fringing?

Yeah, it wasn't 'bow' or 'low' either.

That fly looks like it's got a few dud pixels, by the way.

I did mean 'wow' as well as 'how', by the way. It kinda went without saying. Well, I did utter a word, but it wasn't 'wow'. I've been in trouble from the mods before, so I won't tell you what it was.

I'm not sure I've understood this question right, but if you mean:

"Is the actual subject range, with digital, over which the circle of confusion is a given size or smaller, smaller than it was with film?"

.. then no, it isn't.

The light rays heading in through the front, getting focused and then chopped off by the iris, have no idea whether they're heading for film or a sensor, so the amount they'll be spread out by the time they hit it will be the same, all other things being equal.

However, there is the issue of magnification.

Using the same lens, if you want the whole frame, and enlarge to the same size print, then the amount of enlargement an APS-C 'negative' will get, is larger than the amount of enlargement a 35mm negative will get, so any circles of confusion will be bigger on the print. So you might think there's less depth of field.

But the resulting pictures are not the same. To truly compare 35mm film with digital, you have to crop the inner APS-C-sized portion out of the 35mm negative, and enlarge that to the same size as you did with the digital image.

Then of course, there's no difference.

I suppose, if you used different lenses that were nevertheless hypothetically identically sharp (whatever that means), say a 300mm on digital, and a 450mm on film, and took identical pictures, then the digital negative will need more enlargement than the film negative to get a print the same size.

I have no idea whether the differences are visible though - I never had good enough lenses, and I never enlarged things that much in my film days.

How?

This is mine, at 10mm.

[IMG]http://i162.photobucket.com/albums/t244/chris5gd/Technical/IMGP9460-sm.jpg[/IMG]

It is very noticeable, I agree. Still, it's very easy to fix, and it's an awful lot cheaper than the 12-24. I haven't tried the Sigma at 12mm for comparison.