[BruceBanner]

I've owned the K-1 a few weeks now, I can't help get the feeling that with the same lens I have been using it since day one are performing worse in regards to auto focusing than what I'm pulling from them now. Is it possible for a camera's autofocus ability to get worse over time?
I have a few old manual A series lens, and when shooting manual focus I am very pleased with the results, shooting using the spot focus (where you move the square around the grid and choose the focus point) of late seems to be missing the focus point a lot. I can't tell if it's bad me or camera, I think it's just me, but I thought I'd mention it. (i'm aware of the 3 middle points being better than the rest of the grid, but still even auto focusing centre seems off more often than not...)

Can advice be given on how to customize focus for certain lens (correct sharpness etc)? I'm reading the K-1 ebook currently and it touched upon the K-1's ability to do that.

Cheers,

Bruce

[Adam]

What lenses are you using?

[BruceBanner]

I'm using the 100mm 2.8 F DA WR, the DA 3.2 21mm and the DA 50mm 1.8.

I really can't get the 3.2 21mm to perform well at all, partially i think because it kinda has to be crop mode to be usable, but I get better clarity and results from the 100mm when using in scenarios I didn't think it would do well in (such as panorama's and landscape). But even with the 100mm of late, I'm missing the focus in the centre of some shots, I'm thinking perhaps I need to do a real controlled setup to test each lens and see if an adjustment (in the K-1) needs to be made. I'll likely sell the 3.2 21mm anyway as it's not full frame.

Before I investigate further I just thought I'd post about it, see if it's 'a thing', or whether 9/10 times it's the user at fault.

[pschlute]

I had the same worries when i first got the K1. After a proper focus test and adjustment for each individual lens, it never misses. I needed between +7/+10 on all my lenses

Download a focus test chart and print it.....http://www.kscameraclub.org/docs/pdfs/focus_test_chart_edited.pdf

Camera on tripod, centre spot AF point only. Widest aperture. Before conducting each test get the lens to focus at infinity or closest point (a hand in front of the lens works quite well), so you are making the camera work each time to focus. You need to do about 10 tests with each lens after making any focus adjustment. A single test may not work...you will be surprised how varied the results of a focus lock can be...look at the barrel markings each time.

Also try to do the tests under natural light if you can.

You can do all the adjustments without having to view the photos on your computer, just zoom on ezch image....10x is 100% view

[BruceBanner]

I had the same worries when i first got the K1. After a proper focus test and adjustment for each individual lens, it never misses. I needed between +7/+10 on all my lenses

Download a focus test chart and print it.....http://www.kscameraclub.org/docs/pdfs/focus_test_chart_edited.pdf

Camera on tripod, centre spot AF point only. Widest aperture. Before conducting each test get the lens to focus at infinity or closest point (a hand in front of the lens works quite well), so you are making the camera work each time to focus. You need to do about 10 tests with each lens after making any focus adjustment. A single test may not work...you will be surprised how varied the results of a focus lock can be...look at the barrel markings each time.

Also try to do the tests under natural light if you can.

You can do all the adjustments without having to view the photos on your computer, just zoom on ezch image....10x is 100% view

Thank you, I will give this a try.

[clackers]

Thank you, I will give this a try.

If you haven't calibrated your lenses yet, you need to set aside an evening or afternoon and just do them all.

[Tatouzou]

I dont have K1 but i have also noticed on K3 that AF fine calibration done last year is not good any more and may need a different adjustment on some lenses.
This happened mostly with pancake primes, like DA40 XS, DA21 and DA70 limited.
They are both very sharp and probably more prone to AF errors as they have a very short focus throw.
On the other hand, none of my zooms never needed AF adjustment.

[BruceBanner]

If you haven't calibrated your lenses yet, you need to set aside an evening or afternoon and just do them all.

Ok.

With the K-50, when zooming in and examining quality etc, I figured if something was off it was due to the camera or me, but with the K-1 the cropping ability is extremely good, if you experience a few shots that crop well, but another series of pics don't, and the focus doesn't really seem to be anywhere in particular, and you check the settings and think to yourself that there is no good reason why these shots came out bad... it made me wonder.
I never knew about fine tuning lenses 'til touching upon the subject in the eBook (bought from here in Pentax Forums). The author seemed to think it was probably more user error and not something really needed but it got me curious...

I wonder how it works, does it mean you now have to use your five custom user modes for specific lenses? That would be a shame, as currently my user modes are reserved for shooting parameters/scenes, such as 'Portrait', 'Fast', 'Landscape' etc (and at the time I choose which lens best suits this purpose). I doubt the K-1 has the ability to detect a certain lens connected to it and automatically apply the focus tweaks?

Anyway, yeh, in the near future I will set aside some time and investigate this more and use a focus sheet.

I dont have K1 but i have also noticed on K3 that AF fine calibration done last year is not good any more and may need a different adjustment on some lenses.
This happened mostly with pancake primes, like DA40 XS, DA21 and DA70 limited.
They are both very sharp and probably more prone to AF errors as they have a very short focus throw.
On the other hand, none of my zooms never needed AF adjustment.

Interesting you mention the DA21 as that is the worst performing lens on the K-1 I have. It's quality is as bad as the stock 18-55mm lens that came with my K-50, I have yet to shoot a decent picture with it. Now that it's connected to my K-1 I really feel as tho any flaw is massively highlighted because the K-1 is that good. When trialling out the DA 50mm 1.8 and DFA 100mm 2.8 on the K-1, the improvement was massively noticeable straight away, I have yet to experience that with the DA 21.

[clackers]

I never knew about fine tuning lenses 'til touching upon the subject in the eBook (bought from here in Pentax Forums). The author seemed to think it was probably more user error and not something really needed but it got me curious....

Yeah, autofocus of course relies on all the elements in a lens being properly aligned in the factory, and the AF module in the camera being properly aligned, too.

Roger Cicala of Lens Rentals has shown the performances of the lenses his company owns follow Bell curves ... you can buy a bad one, and you need to ask for an exchange or return, and this might be difficult if you don't test straight away. The seller might say, 'You used it for a year without saying there was a problem. Maybe there is, or maybe you've decided you don't need it anymore and instead of selling it, you're asking for a refund. Or, perhaps you dropped it and shook something loose."

As for the camera, you bought well with Pentax.

All models allow adjustment of the camera's AF module. You need to go high-end in Nikon for that. The most common Nikons need to go back to a service centre.

Here's an article from this forum:

Front- and Back-Focus - Articles and Tips | PentaxForums.com

If you find that +10 or -10 isn't enough for various lenses, the body itself may have to go back to the dealer - the module is too far out of alignment for minor correction, it will have to be taken out and put back in again. Obviously, do this during the warranty period for free, never 'put up with it'. Our dollars are too precious to be wasted.

[pschlute]

I wonder how it works, does it mean you now have to use your five custom user modes for specific lenses? That would be a shame, as currently my user modes are reserved for shooting parameters/scenes, such as 'Portrait', 'Fast', 'Landscape' etc (and at the time I choose which lens best suits this purpose). I doubt the K-1 has the ability to detect a certain lens connected to it and automatically apply the focus tweaks?
.

Fine autofocus adjustment has nothing to do with the USER modes.

The K1 will know exactly which lens you have attached to it, and can store up to 20 individual lens adjustments. See page 54 of the manual.

[BruceBanner]

Fine autofocus adjustment has nothing to do with the USER modes.

The K1 will know exactly which lens you have attached to it, and can store up to 20 individual lens adjustments. See page 54 of the manual.

That is fantastic news, seems like money well spent

[rrstuff]

I've owned the K-1 a few weeks now, I can't help get the feeling that with the same lens I have been using it since day one are performing worse in regards to auto focusing than what I'm pulling from them now. Is it possible for a camera's autofocus ability to get worse over time?
Bruce

Were you changing lenses in dusty conditions?

If you have dust on the AF sensor or the mirror, it could refract the light and throw the focus sensor consistently. Do a test and see if some points work and others don't. Then take a look at the mirror and AF sensor and see if you see any dirt on it. Similar thing happened to me on a k-3 where at some point off center focus points didn't work very well. I later found out some of the off-center AF points work well other don't. Then I read that dirt on the sensor could be a reason, after long long research. I will be cleaning the camera this weekend to confirm this hypothesis.

[Tatouzou]

Ok.

With the K-50, when zooming in and examining quality etc, I figured if something was off it was due to the camera or me, but with the K-1 the cropping ability is extremely good, if you experience a few shots that crop well, but another series of pics don't, and the focus doesn't really seem to be anywhere in particular, and you check the settings and think to yourself that there is no good reason why these shots came out bad... it made me wonder.
I never knew about fine tuning lenses 'til touching upon the subject in the eBook (bought from here in Pentax Forums). The author seemed to think it was probably more user error and not something really needed but it got me curious...

I wonder how it works, does it mean you now have to use your five custom user modes for specific lenses? That would be a shame, as currently my user modes are reserved for shooting parameters/scenes, such as 'Portrait', 'Fast', 'Landscape' etc (and at the time I choose which lens best suits this purpose). I doubt the K-1 has the ability to detect a certain lens connected to it and automatically apply the focus tweaks?

Anyway, yeh, in the near future I will set aside some time and investigate this more and use a focus sheet.



Interesting you mention the DA21 as that is the worst performing lens on the K-1 I have. It's quality is as bad as the stock 18-55mm lens that came with my K-50, I have yet to shoot a decent picture with it. Now that it's connected to my K-1 I really feel as tho any flaw is massively highlighted because the K-1 is that good. When trialling out the DA 50mm 1.8 and DFA 100mm 2.8 on the K-1, the improvement was massively noticeable straight away, I have yet to experience that with the DA 21.

After re-doing AF fine tuning on K3, AF is OK on central AF point but needs to be careful when using any off center point, which makes it a little tricky at f3.2-5.6 because center AF + recompose is not very good too because of the field curvature.
Where this lenses shines is its handling of contrast and colours. It is very resistant to flare and ,stopped down f5.6/f8 it is very sharp.

[BruceBanner]

I had the same worries when i first got the K1. After a proper focus test and adjustment for each individual lens, it never misses. I needed between +7/+10 on all my lenses

Download a focus test chart and print it.....http://www.kscameraclub.org/docs/pdfs/focus_test_chart_edited.pdf

Camera on tripod, centre spot AF point only. Widest aperture. Before conducting each test get the lens to focus at infinity or closest point (a hand in front of the lens works quite well), so you are making the camera work each time to focus. You need to do about 10 tests with each lens after making any focus adjustment. A single test may not work...you will be surprised how varied the results of a focus lock can be...look at the barrel markings each time.

Also try to do the tests under natural light if you can.

You can do all the adjustments without having to view the photos on your computer, just zoom on ezch image....10x is 100% view

Ok, so I'm just getting round to doing this now. I'm wondering if I can seek further advice.

I dled the pdf file, printed page 15, mounted on cardboard and angled at 45 degrees as per instructions. I first noted that 45 degrees seemed a tad steep, other videos for AF testing seemed to have them shallower, perhaps half that?

I kept the tripod and camera fixed and changed the three lenses (that I have that AF), first was the DA F 100mm 2.8, second the DA 50mm 1.8, third the DA 21mm 3.2. I believe I adhered to the test protocols for each lens, that is to use a single centre spot for focus and to shoot wide as possible (so 2.8, 1.8 and 3.2 respectively).

Please see below for the results;

100mm 2.8


50mm 1.8


21mm 3.2 (shot full frame hence vignetting)




Now the first thing that comes to mind is, because I am using a fixed position for the camera its harder to critic the lenses? Should I be pushing the 21mm and 50mm lens closer to the sheet so that the piece of paper starts to occupy about the same visual representation as what I have with the 100mm? For example, of course zooming in with the DA21mm shot at 3.2 towards the black focus line is going to look garbled and awful, i'm having to crop far more to see that area!

Having said that, i did recall reading somewhere that shooting the DA21 at 3.2 is a pretty miserable experience, f8 is far better, and i did take a shot at f8 and can report that the sharpness is far better.

If i perform more of these tests, and I believe above the focus line appears more in focus, then which direction am I moving the +/- 10 scale in teh camera to compensate for this?

Basically I'm just looking for further guidance in performing these tests better before continuing and making any adjustments (if any are even required).

EDIT: I realise the images above are fixed, and don't become clickable links to the source, if you click my flickr signature they should be the first pics in my photostream (thereby allowing you to view the original files and zoom around)

[pschlute]

You should increase the distance from camera to chart as you use a longer lens. The important thing is to have the spot focus point cover the "focus here" point exactly.

Looking at your Flickr pictures at 100% I can see:

The 21mm failed to achieve focus at all. You need to redo the tests.

The 50mm is still exhibitting slight backfocus

The 100mm is exhibitting very very slight backfocus, not that you would notice it in practice.

The important thing about this test is that you are able to do say 10 tests with the same lens and get a good percentage at correct focus (not front or back focus). You will not get 10 correct ones....autofocus is not that good ! Only use widest aperture, there is no point in stopping down. + adjustment corrects backfocus