[Paul the Sunman]

I am really struggling to get decent images from my SMC DA 15 Limited on the K3iii. This was always a good lens for me on the K5 and K3, but I haven't used it much for a few years, since I've most been shooting FF, and APS-C only with the 150-450.

I tried doing an AF fine adjustment. This shot of a focus chart is at f/4 with AF +5. The focus is very poor in the top right and bottom left, which I had not noticed before on other cameras. Is the lens de-centred, and if so is there anything that can be done about it?



I cannot see any fungus or other issues with the lens.

---

Last edited by Paul the Sunman; 06-14-2021 at 02:41 PM.

[bwgv001]

maybe similar to this:
RiceHigh's Pentax Blog: Image Corner Blur Issue with K-3 III (Head-to-head Comparison with KP)

[mikesbike]

I would suggest a retest on another camera. If there is a decentering, there are better ways to check for that, but it would be the same regardless of which camera.

[stevebrot]

I am really struggling to get decent images from my SMC DA 15 Limited on the K3iii. This was always a good lens for me on the K5 and K3, but I haven't used it much for a few years, since I've most been shooting FF, and APS-C only with the 150-450.

I tried doing an AF fine adjustment. This shot of a focus chart is at f/4 with AF +5. The focus is very poor in the top right and bottom left, which I had not noticed before on other cameras. Is the lens de-centred, and if so is there anything that can be done about it?



I cannot see any fungus or other issues with the lens.

If the lens is decentered, there are easier ways to test for it than with a slant target. The method I use is similar to that suggested by an article here at Pentax Forums:

https://www.pentaxforums.com/articles/photo-articles/how-to-check-your-lens-...centering.html

As for the AF performance, an easy check is to see if you do better with manual focus using magnified live view. Skip the slant test in the example photo and use an easy real-world subject. The easiest for the PDAF system is a vertical black bar at more or less right angles to the lens axis. With the DA 15, I would aim at something about meter away.


Steve

[UncleVanya]

I am really struggling to get decent images from my SMC DA 15 Limited on the K3iii.

Double check your settings and make sure horizon correction is off. There’s apparently a repeatable problem with self timer (2 sec and 12 sec) tripod shots using horizon correction on the k3iii. I don’t have a k3iii to test with so I can’t personally attest to the replicability.

K-3 III bug in 2s self timer mode - Source of fuzzyness - PentaxForums.com

EDIT: Paul posted new info that points to a lens problem as this is seen with this lens on his original k3 in addition to the new k3iii.

[kinkindoll]

My DA20-40mm has the focus problem to infinity on my K-3iii, still no idea why it happened. For your problem, I would suggest to use LV to shoot a photo on the focus chart, then adjust the AF fine adjustment to match with the result of liveview in lightroom or photoshop

[ofer4]

Hi, Paul. It looks to me like you're just not square to the target image. Try aligning carefully first and see if the effect goes away.

[Paul the Sunman]

Hi, Paul. It looks to me like you're just not square to the target image. Try aligning carefully first and see if the effect goes away.

No, definitely square on. Repeated with a brick wall, with the same result. The right side (especially upper right) is always well out of focus if the centre is in focus. Tested on both K3 and K3iii.

[Grimmus]

No, definitely square on. Repeated with a brick wall, with the same result. The right side (especially upper right) is always well out of focus if the centre is in focus. Tested on both K3 and K3iii.

Hi Paul, to clarify, you have repeated the test with both the K-3 and K-3 3 on a brick wall ...... and the results are the same for both cameras? Or the results are bad only for the K-3 3 as per the indication in your original post?

[Paul the Sunman]

Hi Paul, to clarify, you have repeated the test with both the K-3 and K-3 3 on a brick wall ...... and the results are the same for both cameras? Or the results are bad only for the K-3 3 as per the indication in your original post?

Yes, basically the same, independent of camera. Face on the a brick wall, the right side is very weak on both K3 and K3iii. So I'm pretty sure the lens is the issue. I will take it to CRK for repair.

[UncleVanya]

Yes, basically the same, independent of camera. Face on the a brick wall, the right side is very weak on both K3 and K3iii. So I'm pretty sure the lens is the issue. I will take it to CRK for repair.

Ah! Well sorry to hear that. It does seem to explain the issue as lens related.

[jumbleview]

Yes, basically the same, independent of camera. Face on the a brick wall, the right side is very weak on both K3 and K3iii. So I'm pretty sure the lens is the issue. I will take it to CRK for repair.

Probably better to change post title. It sounds misleading now.

[kiwi_jono]

While not big, I have seen differences between cameras (K-x, K-5, K-01, K-1 crop) testing side by side and I think for me it was coming down to couple of things:
1. I have noticed that the position of the sensor (moves to different positions while in SR or composition adjust) can have an impact on apparent sharpness across the frame. With some lenses and SR on, I can take a shot of a uniform subject and see small differences in corner sharpness (even with same focus and aperture). I'm now having trouble find an example (apparently I have deleted them!) but there are differences sometimes - particularly with some lenses (for example DA 15 or Sigma 10-20), where they vignette not far out of APS-C frame. Conversely if you use a good FF lens on a crop camera (or K-1 in crop mode), then images seem more consistent (as I would expect).
2. Alignment of the camera mount varies very slightly between different cameras (probably even between different copies). A lenses that looks a little bit out of center on one camera can look Ok on another. Pretty small differences though.

[mikesbike]

Another way I've found to be effective is to find a subject in good lighting at some distance that is solid with good sharp edge relief and good detail. An ornate street lamp is one example, framed against the sky. Then, shooting from the same position, take one shot dead center and check to verify focus and sharpness of edges and of detail. Then immediately under the same lighting, etc take other shots but move the camera slightly to place the object near top and then bottom edges, then near right and left edges. All shots wide open. Can do corners as well.

Repeat to assure focus accuracy.

One thing to be aware of is this lens does exhibit significant field curvature, and of course, lenses in general will not be as good at edges and even more true in corners, but these differences should be pretty much the same on either side.

[Paul the Sunman]

So, I took the lens to CRK in Melbourne for assessment and repair. They found it was indeed decentred, but could not fix it there. It has to be sent to Japan. It may take 6 weeks and cost up to Aus$400, which would be about half the price of a new one. I said to go ahead and send it.