Originally posted by Shanti where did you buy your 12-24? I just ordered one from B&H so hope it is good..hassle with overseas sales to return? was it something that just needed some adjusting? can you explain defect more,so I can test mine when it comes? Thanks
I bought the last two of my 12-24s from B&H, but I wouldn't worry. They are great about accepting the return of a defective lenses. This is why I like them--they stand behind what they sell and make it easy to return a lemon. The problem with the first two copies I tried was severe decentering, which cannot be fixed through an "adjustment". In my experience the only thing to do when you get a badly decentered lens is to return the lens and try a new copy (either from the same retailer or another retailer that you trust).
It is quite easy to test for decentering. I find a scene with trees and buildings along a distant horizon, where the scene stretches across the field of view. I then focus on the center of the scene, and take a number of shots, bracketing the focus (in other words, adjusting the focus slightly). With a severely decentered lens you will find the following: the shots that are sharp in the center will also be sharp on one side, but will be very fuzzy (out of focus) on the opposite side. For example, you may find that the center and right side are sharp, but that the left side is out of focus. When reviewing the bracketed shots with other focus points you may find some shots that are in focus on the left side, but these would then be out of focus in the center and on the right side. You can use the same approach to test the corners. Simply tilt the camera so that the horizon passes from one corner, through the center of the image, and to the opposite corner. Tilt the camera in the other direction to test the other two corners. With a zoom lens this should be tested at a number of focal lengths (for example, 12mm, 15mm, 20mm, and 24mm). Sometimes a lens is acceptable at one end of the zoom range and really far off at the other end.
To be really thorough you may also want to test the lens at shorter distances, although be careful to keep things parallel!! For example, you may want to shoot a brick wall, but if the camera's sensor is not exactly parallel to the wall you can get a "false positive". The images will seem to indicate a decentered lens, but the problem is that you have not aligned the camera properly. This is why I like to use a distant view. The results will not be biased by misalignment of the camera if the horizon is at "infinity". If you want to test whether your horizon is in fact "at infinity" you can shoot the same scene with the camera upside down. If the image was soft on the left side with the original orientation it will still be soft on the left side when the image is inverted.
Before you do any of this I suppose you should be confident that your camera does not have defective sensor alignment. If many of your lenses show the same pattern of "decentering" (for example, soft lower-right corners) then the problem may be with your camera rather than the lens. You can rule out the camera as a cause if you have more than one camera with which you can run the test, or if you have tested a lot of lenses that do not show the same pattern.
Finally, I suppose I should mention that this is not just a problem with Pentax lenses. I have owned and tested more Nikon lenses than Pentax lenses, and the Nikon lenses are just as prone to decentering issues. Based on my experience, prime lenses are less likely to suffer from decentering problems, but I have seen a few primes that were horribly decentered, so they should all be tested. I have not had a problem with Pentax Limited lenses, but I'm sure that I have simply been lucky because others have received decentered limiteds. My DA 70mm Limited arrives today. The first thing I will do is to test it for decentering.
Dan