[jpzk]

I have just received this new DA 15 Ltd (almost new) lens and tested it with roughly 100 pics from landscape to close ups and I find:

the shots taken at close range, in this case "artificial flowers" show perfect focus at about 2 feet away. I also took some of very small print, at close range ... all perfectly focused.
the shots taken during landscape sessions do not look very good: almost slightly OOF overall.

There is absolutely no way I can get remotely close to the quality of the photos from this thread: https://www.pentaxforums.com/forums/lens-clubs/86234-15mm-limited-controls-my-mind-club.html

Since I dumped most of the pics, I don't have them to show here.

My question: what is the best way to "fine tune" the AF so that it is consistent at both close range and "landscape" mode.

I haven't touched the fine AF adjustment for BF/FF yet, nor didI think this was necessary because the close up shots were bang on.

JP

P.S.: I haven't tested it on my K5 yet. But that shouldn't be an issue I think.

---

Last edited by jpzk; 10-16-2011 at 08:22 PM.

[Catalana]

Without seeing any shots and exif it is hard to know what your lens is doing, but I would close in on the landscape problem by taking several shots in manual focus mode at varying "known" distances up to infinity
and repeating the series with the "same" focal points at varying f-stops. You should be able to see the differences in the choice of f-stop and also find out if the lens focuses at infinity. Then I would repeat the same
exercise with spot auto-focus (i.e., the center focus point), keeping in mind, that autofocus may not always focus on the exact spot you think it is focusing. You can also check the distance scale on the lens after you have auto-focused
and see if it is in the ball park. Little tests like that, along with taking notes can go a long way in solving some problems.

[v5planet]

I had this exact same problem when I first got my DA15 as well. It seemed to resolve itself as I used it more. ;-)

In all honesty, I think I made a very slight AF microadjustment, but the perceived sharpness in landscape shots seems to have a lot to do with lighting and whether you have objects that extend towards the foreground (an increased sense of depth seems to enhance how sharp things look overall...somehow). Also... tripod tripod tripod. I tend to shoot landscapes around f/10 -- that helps. As does PP. Anyway, don't despair, the lens is probably fine.

[imtheguy]

Sorry such a cool lens is aggrevating you. As far as adjustment, the 15 and the 10-17 fisheye are the only ones of my 15 or so lenses I did not bother to calibrate since the DOF is huge. As stated above, how about linking to some big samples. I can only imagine that if its sharp up close but never at distance I would check for complete focus at infinity or just swap for a different 15. Of course, I am assuming that your OOF distance shots were actually focused at distance.....you can test that manually of course. Good luck, let us know.

[v5planet]

Sorry such a cool lens is aggrevating you. As far as adjustment, the 15 and the 10-17 fisheye are the only ones of my 15 or so lenses I did not bother to calibrate since the DOF is huge. As stated above, how about linking to some big samples. I can only imagine that if its sharp up close but never at distance I would check for complete focus at infinity or just swap for a different 15. Of course, I am assuming that your OOF distance shots were actually focused at distance.....you can test that manually of course. Good luck, let us know.

Yes, a way to test this would be to pull up live view and zoom in fully, manually focusing to the highest degree of precision you can and then snapping off a few shots.

Also, remember that vibrations from the K-7's shutter action allegedly can create a bit of fine blur in wide angle shots at shutter lengths from 1/100 - 1/200 or so. I would really recommend setting up on a tripod and using mirror lock up for the purpose of these tests.

[reeftool]

I have on a couple of occasions mounted the DA 15 on my camera after using a manual lens and just started firing away, never noticing that the camera wasn't focusing (senior moment, I know nobody else has ever done that). Probably 80% of the shots came out perfectly and I was able to salvage many of the others in PP. The DOF is huge with that lens.

[jpzk]

Thnaks a whole bunch guys!

OK, so I am going to proceed with a few tests as described above, including manual focus, AF, using the tripod and MU, etc ...

I think I won't bother with charts as they often led me to believe that the focus was off while it wasn't, on some other lenses.

I took a few this afternoon ... I'll have to upload them to see what they look like.

Cheers.

JP

[wlachan]

Just do AF calibration with near infinity target. The reason is that the focus won't be affected as much as the focus distance decreases. This is also the reason AF charts don't work.

[jpzk]

Just do AF calibration with near infinity target. The reason is that the focus won't be affected as much as the focus distance decreases. This is also the reason AF charts don't work.

So, you'd go manual focusing for that? And, how "close" to infinity would you go ?
Seems like an interesting way to try calibrating this lens.

Now, I must admit that this same lens doesn't need any adjusting on the K5 vs. the K7.
But because I want to carry one camera for my usual "long shots" - K5 + DA*300/4 - and one "smaller kit" for other-than-telephoto stuff, the K7 has to work properly with this DA 15mm.

All other lenses I use with the K7 work great, including my DA*16-50 ... so I hope to be able to tweak this to work.

Thanks for the hint, which I will try.

JP

[wlachan]

I did the same AF calibration with my DA14 & K-m. 2-3mm away from the infinity mark will do. You may check out the following post for detail.

Ditto: Pentax SLR Talk Forum: Digital Photography Review

[jpzk]

Thanks!

I've checked the post from that link ... rather useless for me because I can't figure out what it says in there. Sorry, my fault.

I'll try to use your suggestion. Hopefully that'll work

JP