[dijkx]

My first Kr went back because of heavy frontfocusing. I did some testing with my second Kr.
I did the same test as earlier, I placed a ruler at 45 degrees and focused on a piece of wood next to the ruler wich is not at a angle, the camera was at twice the focus lenght (in this case 1.40 mm) This test was advised by Sigma Benelux.
I did one test at daylight and one with tungsten light, as you can see on the pictures. the camera is backfocusing at daylight and frontfocusing with tungsten light. I did the same test with my tamron 90 mm, same results.
The pictures are taken with my Sigma 70-200 HSM II at 70mm at F2.8

Are there more Kr owners with this issue ? and can it be resolved in a firmware upgrade? What I can see on the internet I'm not the only person with a FF problem.

[lock2nl]

Where did you put the focus on ? Left or right cross ?

[dijkx]

Where did you put the focus on ? Left or right cross ?

On the right cross

[lock2nl]

It has nine cross-type sensors with a higher sensitivity. These are wider and thicker then normal ones, and are apt to focus on things you did not intend to. (The Sony a700 is a notorious example of this issue). I would blank out the right cross, focus on the left one and bring the ruler outside the range of the central sensor. That way you are sure it focuses on the bar, not on the ruler. If you are sure it focused right, check out the live view result. If the results are fine with LV, I think you should consider sending this one in or ask for new one at the dealer.

[Hypocorism]

On the right cross

Tried this method?
Nikon D70 Focus Chart

[ requires d/l of http://focustestchart.com/focus21.pdf ]

.R.

[lock2nl]

Forget that one. The Af test I received from Sony and Pentax were all without a 45degree ruler to focus on. You need something parellel to the sensor to focus on. Actually, sony firsthanded me over the moiré test to start with. Really informative. Or should I say confirmative ?

I have used this one as well btw. But results are too inconsistant to make much of it.


lock

[rawr]

I did some testing with my second Kr.

Your testing procedure could probably be better. Otherwise your conclusions about how your AF is performing may be flawed.

This is how Pentax did AF testing in the K100D days. You may like to adopt the same procedure - following all the steps - and see what your results are. Note that the test should be done with the lens wide open, for minimum possible DOF, and they recommend using the FA 1.4:




The revised v2 confirmation chart that replaced the chart used in the diagram above.
Print it out to about A4 size, cut out the chart and scale, and set it up in the way indicated in the diagram above:


The original v1 confirmation chart:


Of course, when doing AF testing, you should have NO polarising, UV or any filter of any sort on your lens.

As Canon and others have suggested, it may also be useful to perform some AF tests at a variety of distances (5m, 10m, 20m etc), perhaps even outdoors. If you are a landscape or sports photographer, AF tests done indoors at 1m may not tell you anything useful about how the AF of your system is performing.

FYI, these are useful reading about AF:

LensRentals.com - How Autofocus (Often) Works.
and
LensRentals.com - "This lens is soft" and other myths

[dijkx]

p { margin-bottom: 0.21cm; } rawr: Thanks, I'm going to try your method, I haven't seen that one before. The method I used was recommended by Sigma Benelux. The most important thing in their test was not to focus on the 45 degrees ruler but on a object parallel to the sensor. (that's what I did with the piece op wood)
I will try to make some time to test your method, I’ll will post the results on this forum

[rawr]

I will try to make some time to test your method,

Actually, it's not my method It's from the official Pentax K100D Service Manual.

[dijkx]

Actually, it's not my method It's from the official Pentax K100D Service Manual.

I did some testing with "your" method, I could not test at daytime because it's already dark when I'm home but did some testing with two different light sources, halogen and LED. I was surprised to see the differences in focus and I'm not very happy with the results. I did the same testing with my K10d and the results were much more consistent.
Front focus with halogen light and backfocus with LED light I wonder how I can ever make a sharp photo in tungsten light ??
Does anyone has the same experience with his Kr ?

[DonP]

Under daylight illumination my k-r focuses great. It was much better out of the box than my k-x. I do have a small amount of front focus under tungsten lighting. It is noticeable in focus testing and also on normal subjects - but only at relatively close range. The further the focus point is from the camera, the more the depth of field covers the error. My fastest glass is an FA50/1.4 and I don't use that much at close range so the slight front focus doesn't cause me to lose any sleep.

I couldn't hazard a guess as to the color temperature of either of the lenses you just tested with. But I'm pretty sure the chances I'll ever use LED lighting for a subject, other than maybe a Christmas tree, are pretty remote. Dial in your fine focus some day when you have daylight to use and then see what kind of error you get under tungsten lighting and whether it's noticeable in the type of indoor shooting you do. Your k-r is a fine camera - it's not your old K10D when it comes to focusing under artificial illumination. But it's a great tool, nonetheless. Try to enjoy it.

[lock2nl]

Well, I tend to disagree with a 'let go' approach. If you are not comfortable with a camera, you should either exchange it or get it fixed (if that will ever help, of course).

AF modules are pretty complex. A primary piece of glass that lets through the light, a secondary one to reflect the ligh to the AF module, and third reflection mirror to point the light to the actual sensors. Any malalignment of these mirrors may lead to different beams, particularly if the light has varying wavelengths. My suggestion is to have the internal parts checked.

lock

[Deimos]

use the Tungsten white balance setting when focusing under tungsten makes a noticeable difference in the focus, may still need to af adjust a bit

[woof]

use the Tungsten white balance setting when focusing under tungsten makes a noticeable difference in the focus, may still need to af adjust a bit

This is just what I would expect... this is the "hook" that Pentax would have to improve AF adjust under tungsten. At a minimum, I would think they could make a global AF adjustment based on setting WB to tungsten. Might not be perfect, but it would be a step in the right direction.

I've purchased one of these. It's going to be used chiefly out-of-doors, so I am not too concerned, but I will probably do a bit of testing using the LensAlign system which is simply a commercial version of the idea presented here. Will get round to posting results.

woof!

[Deimos]

see this thread:
https://www.pentaxforums.com/forums/pentax-k-r-forum/123165-my-k-r-auto-focus...ml?src=subsct2