K10 focus check

I have adjusted my K10 twice. Initially after approx 1000 actuations and then again after approx 7500 actuations. This would seem to support the view that a certain amount of 'bedding in' has occured, though I accept this is far from proven.

As for the 50 mm, mine was fine originally, but my FA 35 , FA31 and DA18-55 DA 50-200 kit lenses were all slightly backfocussing. I made the appropriate adjustment (40um), which resulted in the FA 50 slightly front focussing.

I adjusted for the second time when I purchased a DA 16-45. I checked the same lenses and found I needed far less adjustment (10um). As a result I now detect no front focussing on the FA50.

Steve

Very interesting information, thanks, Steve. I did wonder if it was possible for the adjustment to itself require adjustment, and from what you say, perhaps it does.

If I get the chance I might recheck again at the weekend. I will keep you posted.

Steve

Hi Guys.

This is really a very interesting (to me anyway) thread.

Could someone tell me if the adjustments are only going to affect AF. In other words if I manually focus the results should not change before and after adjustment?

Thanks
geoff

I don't think the way the AF system works out whether the shot is in focus is any different whether you focus it manually, or allow the screw drive (or lens motor) to focus it.

So if you have the green hexagon focus confirmation, that shows that the AF system thinks the picture is focused.

Given that, I would imagine that any adjustment (and I don't know what it is that's actually adjusted) applies equally to AF and MF lenses.

I'm only guessing though - I don't know much about how it works.

That too is my understanding, but like Chris I have no proof that this is the case.

Regards - Steve

When you focus manually you focus until the image is at its sharpest on the ground glass. The distance from the ground glass and from the sensor have to be the same and fixed, otherwise all shots would be off.

Therefore I'd suppose that the change being made is to the AF calibration so that its focus point coincides with the actual focus point, rather than anything being physically moved.

There will be a few factors involved, such as also the accuracy of the lens manufacture and its ability to stop dead in its tracks at the focus point, rather than over-shooting.

Has anyone tested the focus point from both directions? That is, say when focusing on 5m has it been tried from the lens starting at the closest focus as well as from infinity? Does the lens focus equally well from both directions? If anything I'd expect front focus from one direction and back focus from the other in a fairly consistent manner.

Being able to set an adjustment for individual lenses as on the K20D seems like a good feature to have.

At the severe risk of seeming simplistic can I ask why such adjustments have to be made at all - or is this an advantage of digital over the old film slrs?

Could one say that if any adjustments are required (eg due to less than expected image quality) then this is indicative of a manufacturing fault?

Does the same need arise with those other manufacturers of dslrs?

Thanks again all
Geoff

The answer to the last question is yes, it applies to all marques, but only Pentax and Nikon supply a user solution.

It was a problem in film cameras too, but one of film flatness, and not that well recognised.

The problems in digital are different and also exacerbated by us looking at our images with huge degrees of magnification!

I have just carried out a rather simplistic focusing test that may be of interest.

Using a variety of lenses I focussed on a fixed object approx 3 feet away, first moving from closest focus to when the green hexagon appears, followed by moving from infinity again to when the green hexagon appears. The camera was tripod mounted.

The lenses I used were DA 16-45, DA 18-55, FA 31 ltd, FA 35, FA 50 KA 50 and K 55.

I was interested in the amount of movement I could make to the focus ring while maintaining the green hexagon.
I accept that the measurements were crude and more than a little subjective, but as I repeated the exercise several times, I think that the results can be taken as a reasonable indicator.

DA 16-45 F4 @ 16mm 6mm
DA 16-45 F4 @ 45mm 2mm
DA 18-55 F3.5 @ 18mm 4-5mm
DA 18-55 F5.6 @ 55mm 2-3mm
FA 31 F1.8 3mm
FA 35 F2 2mm
FA 50 F1.4 1-2mm
FA 28-70 F4 @ 28mm 4mm
FA 28-70 F4 @ 70mm 1-2mm
KA 50 F1.7 2mm
K 55 F2 3-4mm


Quality of lens build is obviously a factor, I found the 18-55 the hardest lens to focus manually.

Build quality aside I think we can see 2 effects both of which not surprisingly relate to depth of field i.e. focal length and aperture.

The result with the K55 appears at first to be out of synch with the rest until you remember that this lens has a very long focus throw (3/4 of a turn from close to infinity focus). I would expect a 50mm manual macro lens to throw up a similar result.

As an afterthought I took a KA50 F1.7 and set the aperture to F8. I then fitted it to the K10 but only by about an 1/8 of a turn. In other words the auto aperture control was not engaged. I then carefully repeated the focus test

KA 50 @ F8 3-4mm

My conclusion is that for me in 95% of real life situations the K10 is focusing within acceptable limits and I can let the camera take care of it.
For closeup or portrait work when shooting with large apertures, then I will to switch to manual focus and use the screen and my eyes rather than relying on the green hexagon.

Regards - Steve

Very interesting test, many thanks for sharing the results.

Just a point when doing AF adjustments: you should always move the lens waaaay out of focus before getting the camera to AF I normally shift to infinity before AF-ing on the test target.

The 50/1.4 can be a pain to adjust anyway. It's field isn't flat, and I've found a combination of test target at f/1.4 plus some portrait tests to get the right balance. I've normally found the test target adjustment produces a measure of FF, so somewhere in between is usually good - slightly BF seemed to produce sharper portraits for me. Just a thought

Matt

As John says, very interesting, Steve.

I noticed some time ago, when taking pictures of the moon, that accuracy of focus was incredibly critical - I kept getting unexplained duds, even though I had focus confirmation every time.

As you have, I also noticed that there's a range of a few millimeters of focus ring position over which the green hexagon is displayed.

I found that if I took the picture when the focus ring was right in the middle of that range, I got more shots in focus. Still only about 1 in 3, but a lot better than the 1 in 10 or so I was getting before.

As a matter of curiousity I just carried out a limted test using the same methodology , but this time with my MZ6. I will not document all of the results, but it certainly seems like there is less 'tolerance' with the MZ6.

DA 16-45 F4 @ 24mm 2-3mm of focus movement
DA 16-45 F4 @ 45 1mm
FA 50 F1.4 1mm
K 55 F2 1-2mm.

Maybe the brighter viewfinder of a full frame 35mm is a factor here, allowing a tighter tolerance to be used. I certainly did not notice a trade off with time to lock on focus.

This would support the view that the revised viewfinder of the K7 is an attempt to get back to a tighter focusing tolerance without any perceptable loss of speed.

Regards - Steve

SR is off with all the remote and delayed modes, regardless of the position of the SR switch, so I can't see how SR is at all relevant here.[/quote]Yes, SR is off with 2 sec. delay set, but SR stays on with cable remote without 2 sec. delay (as I found out when I tried this procedure the first time!).

Edit: The 2 sec. delay didn't make any difference to my results in the focussing test (solid tripod I guess :wink. However, I found that the 2 sec. delay is essential in any astro photography.