AF problems Tamron 17-50

Started Apr 23, 2012 | Discussions thread
lister6520 New Member • Posts: 3
Re: AF problems Tamron 17-50

Maybe I'm asking a silly question, but are you determining that it focuses wrong from the photo itself or by observing the difference on the distance scale between the over 5.6 and under 5.6 apertures?

I ask because I have a Tamron 10-24 that behaves in a very similar way on my K-30  but the camera is actually doing the right thing.

The lens does have some considerable focus shift (due to spherical aberration at wide aperture) so such a behaviour is actually required.

I have done some tests and found that at F3.5 the focus point chosen by the camera is indeed the correct one for F3.5, but will appear slightly back focused if I then expose at F8 without refocusing.

If I focus at F8 then the focus is correct if I expose at F8 but then if I expose at F3.5 without refocusing I get a front focused image.

In other words it seems the camera and lens are doing exactly the right thing but if I were to observe the focus ring it would look wrong. The reality seems to be that the camera is correctly correcting for the focus shift.

If I use CDAF on the other hand I get correct focus at wide apertures but at narrow apertures I get back focused images. Of course at the lower apertures there is more depth of field so the error is not so noticeable.  CDAF is said to be better because it 'sees' the same image that will be photographed, but that is only true if you expose at the same aperture as you focus, or if thelens has little or no focus shift.

I do also notice some slight shift on other lenses when changing aperture but it is tiny compared to what I get on  the 10-24.

My guess is that the K-30 has separate focus sensors for wide and narrow apertures, a high accuracy one for wide aperture and another one with a shorter baseline for slow lenses (typically telephotos and long zooms). They then probably took advantage of this to optimise the focus accuracy of lenses suffering from focus shift, since the two sensors can get a good estimate of the focus shift of the lens without actually stopping down the aperture (the F8 sensor, even at full aperture, sees the same light rays it would see if the lens were stopped down to F8)

I don't have the Tamron 18-50 but I am guessing its behaviour might be down to the same reasons.

It would be interesting if someone else having a Tamron 10-24 could check if the same behaviour is exhibited.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow