[jpyykonen]

I've had this lens about three weeks now and I'm really starting to doubt that the lens is a bad copy. I hardly get sharp images even at F8, which I expect to deliver good resolution especially from the lens of this price point. Many cases I'm sure I would have gotten better results with 18-55 WR kit lens, DA 18-250 and especially with Tamron 17-50 (which I all own). I don't even know if the problem is the autofocus, shutter blur or just the bad lens. I haven't had any of these problems with any of my lenses before. This is by far the most expensive of my lenses so I really expect it to be somewhat better than the others (maybe not better than the Tamron in case of sharpness, but at least near the same level). Here are some sample pictures all at F8

Full


100% crop near the center


Full


100% crop near the center


100% crop upper left


Once I happened to have acceptable sharpness. Full


100% crop


I have lot of "bad" samples like this. The sharpness just doesn't seem to be there. Is this just a "characteristic" of a lens or is this really a bad copy and should I send it back? I have tried SR on and SR off, same results. I have tried with shutter speeds 1/200-1/1000 instead of 1/60-1/150 with same "not-so-good" results. Results at F4.0 are about the same, F2.8 being even worse. Sometimes I get better results at F5.6 than with F10. I seem to get better results in closer range than with near to infinity focus. All I can say that the results I get are somewhat inconsistent. And as I previously mentioned, I have had no problems like this with any other lens I own. I have tried the lens also with K-7 and with K-5 with same kind of results.

There is lot to like in this lens, especially the mechanical quality and feel. Only if it would produce good images

[liukaitc]

if the image you get from 16-50 is worse than your kit lens, then you can be pretty sure it is a bad copy.
I think 16-50 is probablay worse than kit lens at f2.8 to f3.5. after f4, it is better than kit lens.

try use manual focus with LV, see what you can get with 100% accurate focus.

[Rondec]

Looks pretty soft to me. My experience is that there is some corner/border sharpness at f2.8, particularly at wide and long ends, but beyond f4, even edges should be pretty sharp. There is definitely some field curvature that can produce softness in some situations as well.

[bwDraco]

Your images are softer than what I get from my DA L 18-55mm lens shooting wide-open. At f/8 from a lens of this caliber, you should be getting tack-sharp output. This certainly does not appear to be normal, especially when this is supposed to be a professional lens. You should exchange the lens or have it replaced it under warranty.

--DragonLord

[PePe]

That does not look right to me. A good copy should have pretty good sharpness in the center. Wide open at 16 mm the edges may be a different story, but even there these pictures are not what they should be.
Just to make sure we are not jumping into conclusions: have you calibrated your AF with this lens?

[bassek]

A good copy of the kit is good, I recall the DA16-50 was outperformed by the kit in a German magazine test...

But in these pics, the 1st one seems to be focused close, thats why the cropped birchtree is soft. The second is taken towards the sun so it is a difficult shot, even with hood attached.
Try to focus manually as liukaitc advised.

Seb.

[jpyykonen]

Thanks everybody for the answers. The problem may very well be the front focusing after all.

Just to make sure we are not jumping into conclusions: have you calibrated your AF with this lens?

I did the AF adjust when I got the lens and I ended up with +8. But it was done with focusing chart inside (in daylight) and with close focusing distance. Today I made some test shots with longer focusing distance and +8 was way too much so the lens was front focusing (as bassek mentioned in the post before; I wonder how I didn't notice the front focusing in the original resolution photos but bassek could tell that in that downsized photo? "The photographers eye" perhaps? :P ) I should have checked the AF calibration before So I tweaked the in camera AF adjust and it got better.

I wish I had more time to shoot with the lens as it produced obviously much better resolution after the more proper AF adjust. I guess the best option may be to send the lens and camera to proper AF calibration as the lens is under a warranty. Somehow I like the lens, I would not want to part with it

[Marc Sabatella]

No, it's impossible that the problem is front focus. That would be an issue only if you carefully forced the camera to focus no one particular tree, and the focus ended up being an inch or so in front of that particular tree. What seems to be going on here is that you didn't carefully control which tree the camera focused on, and one of the crops you posted are of the right tree. That and/or the lens might just be soft overall. Impossible to tell without knowing more about how you selected the focus point, where you aimied that focus point while shooting,whether you recomposed after focusing etc. I am also assuming, BTW, you weren't using an ID filter (Image Degradation; sometimes sold under the misleading abbreviation UV).

[bassek]

Here is my suggestion:

Do the brickwall test, first with AF and then several MF shots with same f-stop.
Then repeat from a 45 degree angle.

Seb

EDIT: Use hood but no filter.

[jpyykonen]

The shots before were taken without ID (UV filter) using center focus point with slight recomposition. The focus point was roughly at the point where I took the crop.

Today I finally got my K-5 which was in service because of sensor stains. Sensor was replaced and I decided to try the DA* 16-50 on K-5. What a difference! No more that kind of "fuzzy" pictures. I haven't time to make any AF adjustments but the results were pretty good right from the start. I managed to find a brick wall also so I did the brick wall test and lens seemed as sharp as I had expected. Good at F2.8 and excellent from F4 forward. Somehow F5.6 seemed sharper than F8 in some photos, but not much. So It seems like at least some of the problem was in the focus accuracy with K-7. I wonder why I had problems with K-7 but not with K-5... K-7 worked fine with Sigma HSM lenses so the problem shouldn't be in the SDM contacts.. But anyway, I'm happy with the lens performance on the K-5. Should shoot more to get more experience with the lens. Here are some from today: