[PocketPixels]

I will soon list some Limited lenses for sale. This is not that post. This is a post about the best way(s) to test and demonstrate the sharpness (or softness) of a lens.

The 35mm Limited I wanted came bundled with two other Limited lenses I already had. So now I have 2x DA 70mm and 2x DA 15mm. I find myself in the unusual position of being able to compare 15mm vs 15mm and 70mm vs 70mm to see if there are any significant differences between them. What I have found is that the biggest difference is not between my two copies of a lens. The biggest difference is between shots taken focusing in "from infinity" and focusing out "from close-up". Both of my 70mm Limited lenses perform well ONLY when they focus "in" from infinity. They deliver poorly-focused results when they focus "out". The difference is so great that it might change my whole photographic technique if this keeps up.

I set up a test target on a sunlit patch of cobblestones near the office. Tripod, 2-second timer with mirror lock-up, sunny day. I had to use a CPL filter due to the midday sun. I don't like introducing that variable, but I took note of the CPL angle. Center focus point, pattern metering. Three shots at f/2.8 to test depth of field, bokeh, and field curvature. Three shots at f/5.6 to test overall sharpness and corner brightness. Before every shot, I defocused the lens and refocused it using the center AF point of my K200D. (I recognize that field curvature is hardly an issue with a 70mm lens, but it's practice for when I test my two 15mm lenses.)

At first, my 70mm Limited really seemed to struggle at f/2.8. I was very disappointed. This is the best of three test shots, each one a fresh AF lock after I focused on my nearby hand:

Poor. Lens is backfocused. I would only sell it with a disclaimer and a discount.

But then, I tried three more shots. This time, I manually set the lens to a distant focal point and then used AF before each shot. This is the best of three such shots:

Very good. Lens is fine.

For a side-by-side comparison:


I did these tests on my lunch break, using my work camera, a K200D. That camera does not have any lens adjustment abilities, so a backfocused lens is no good to me. (I do have a more capable K-S2 at home, but I want lenses that work well on both bodies.) Am I testing things correctly? Any advice?

[UncleVanya]

You can adjust fine focus on the k200d but only with a global adjustment:

K200D autofocus fine tuning. - PentaxForums.com

[bobbotron]

My 70mm focuses very well with my K3, it's been my star lens of 2020.

[caliscouser]

Maybe it is time to upgrade that work camera

When comparing these lenses in order to decide which to keep I'd offer a few other suggestions on what to test.

De-centering, particularly with the 15mm.
Lateral CA
LoCA particularly with the 70mm wide open
Infinity focus

[Unregistered User 8]

Lens is backfocused. I would only sell it with a disclaimer and a discount.

I think you are doing yourself quite a disservice here PocketPixels. How a lens performs on your old K200D camera body is no indication how it will go on my K-3.
Remember, it is the camera your are adjusting with AF tweaks, not the lens. How you decide to pick which lenses to keep is your decision, but I would not sell the others cheaper, because you think they are inferior. Just my $0.02.

[c.a.m]

Am I testing things correctly? Any advice?

I would say that your test set-up looks reasonable for what you're checking. The target is simple and has a high contrast, so it should be unambiguous to the K200D's AF sensor. Not sure of your target distance; I usually use at least 25x focal length.

In my tests, I have not found that a CPL affects focus accuracy.

I can corroborate your experience with direction of focus. I have several lenses, including a DA Limited, that have different focus accuracies depending on whether they are focused from infinity or from close-in. In other words, the optimal AF Fine Adjustment value depends on focus direction. For example, for best AF performance at its optimal AFFA setting, on my DA* 50-135mm I focus from a 'near' position. Most of my AF lenses are not so picky on my K-3 II.

The focus accuracy, or front/back-focus issues, depend on the combination of the lens and the camera. Almost all of my lenses need an AF Fine Adjustment. I agree with @pjv -- a lens that requires an AF fine adjustment is not necessarily substandard. A particular front- or back-focus issue might be specific to your lens/camera system, so another user could find a different dynamic on their camera with the same lens.

I've spent considerable time to calibrate the AF Fine Adjustment precisely for my lenses, and I believe that I've established a rigorous, repeatable procedure. Here's a technical article on my experience with manual focus lenses, which provides some insight into the methodology and several examples where the focus adjustment is different depending on the focus direction: Achieving Better Manual Focus with the Green Hexagon - PentaxForums.com

- Craig

[mikesbike]

Without running precision testing, I have a distinct impression good enough to know that AF with my screw-driven lenses I've used through a number of models, including my old K200D, are now performing with greater speed and accuracy on my KP, compared with use with any model before it, the most recent of which are my K-5 IIs and K-S2. I have always preferred using spot focus for most of my shooting situations. Without more extensive experience using multi-point AF, I cannot offer personal comment on AF improvement in this regard, but I have read positive reports from other KP owners.

[pschlute]

Am I testing things correctly? Any advice?

I echo the comments that the back-focus is not a property of the lens itself, but of the lens/camera combo. No need for you to mention this when selling the lens.

The screwdrive AF mechanics are not as accurate as the more modern counterparts. Also your observations that there is a difference when focussing from infinity versus closest focus distance is one I have encountered many times. PDAF is not a 100% accurate system. When doing a lens calibration on my K1 for example I will do 10 tests starting from infinity then 10 tests starting from closest focus distance. I will settle on a AF/FA setting that agrees with the majority result. I never get 20 identical results, especially with screwdrive lenses.

I would suggest a focus chart which can show FF/BF clearly. Print off the one at the end of this article........http://www.kscameraclub.org/docs/pdfs/focus_test_chart_edited.pdf

[Apet-Sure]

Rest easy. A camera/lens combo needing an AFFA adjustment is by no means unusual, even with a Limited or 'Star' lens. There are an awful lot of ones and zeros that the camera's AF processor has to deal with when it's evaluating a scene. As new camera models are released, their AF algorithms and processors improve. But there will always be some manufacturing variances in lenses and cameras. Maybe an optical engineer can explain why some lenses have a better AF hit rate when focusing in from infinity rather than out from close-up.

Maybe not helpful for your K200D, but you may want to watch the Youtube video about the 'Dot-Tune Method' for AF fine adjustment. I found it interesting and instructive.

You Tube

You Tube

If you decide that one of your 70mm Limiteds is absolute trash and you want to sell it for a song, be sure to PM me. But I'll bet there is nothing wrong with them.

[PocketPixels]

Maybe it is time to upgrade that work camera

When comparing these lenses in order to decide which to keep I'd offer a few other suggestions on what to test.

De-centering, particularly with the 15mm.
Lateral CA
LoCA particularly with the 70mm wide open
Infinity focus

I'd love a new work camera, but truthfully the K200D does fine. For product photography, at least, the lighting setup is more important than the camera. Before I got here, my boss was using a $500 lightbox, two $100 LED spotlights…and his iPhone. And he was getting B, maybe even B+ results.

I would say that your test set-up looks reasonable for what you're checking. The target is simple and has a high contrast, so it should be unambiguous to the K200D's AF sensor. Not sure of your target distance; I usually use at least 25x focal length.

In my tests, I have not found that a CPL affects focus accuracy.

I can corroborate your experience with direction of focus. I have several lenses, including a DA Limited, that have different focus accuracies depending on whether they are focused from infinity or from close-in. In other words, the optimal AF Fine Adjustment value depends on focus direction.

I've spent considerable time to calibrate the AF Fine Adjustment precisely for my lenses, and I believe that I've established a rigorous, repeatable procedure. Here's a technical article on my experience with manual focus lenses: Achieving Better Manual Focus with the Green Hexagon - PentaxForums.com

- Craig

Thank you, Craig. The CPL is a factor only insofar as this is a cheapo resin CPL. If the lunch-break sunshine weren’t so bright, I wouldn’t involve this filter in my lens testing. I’ve just read your article on the Green hexagon, and it quantifies something I’ve often suspected. I do a lot of MF macro shots at work (plumbing parts and pool accessories), and I have a love/hate relationship with that hexagon.

My first post on these forums was a 2011 question about back-focus, so focus has been a concern of mine for 10 years. But only this week did “direction of focus” become something I’ve considered.

Maybe not helpful for your K200D, but you may want to watch the Youtube video about the 'Dot-Tune Method' for AF fine adjustment. I found it interesting and instructive.
DotTune: Autofocus fine tuning in under 5 minutes - YouTube

If you decide that one of your 70mm Limiteds is absolute trash and you want to sell it for a song, be sure to PM me. But I'll bet there is nothing wrong with them.

I’ll check out that video, thanks!