[Photomy]

Just received K10D a few days ago and noticed the red center AF indicator point is not properly centered in the central viewfinder bracket. Do we accept this as the norm? Is this common? Worth sending it back for a replacement?

Thanks for any feedback.

P.S. Just to clarify, I am referring only to the red, small, indicator box. The real focus point might be perfectly centered within the viewfinder. I have read that it is actually a cross in shape. It is hard for me to verify this. Did some testing by focusing on window pains, etc. but still could not really know for sure.

---

Last edited by Photomy; 02-07-2007 at 11:28 AM.

[Dana G]

It makes no difference as long as you know where is and point it at the subject. The actual "in focus" indicator is the green hexagon at the bottom of the viewfinder field.

[Ivan Glisin]

The actual position is not important at all: red indicator dot is not linked to the physical position of the AF sensor at all (red dots get projected in the pentaprism, while AF sensors are under the mirror in the mirror chamber). Red dot just tells you that AF system has successfully acquired maximum contrast in that area (maximum sharpness in most cases, although not all).

But more important thing to remember is this: AF point is NOT an AF point at all! It is an AF area, thus much bigger that that little red dot. Draw a black dot on a blank sheet of paper and see for yourself how big AF area is: try focusing while moving AF "point" away from the black dot and I guarantee you'll be surprised.

Failure to understand this often results in unexpected and apparently wrong AF operation.

[RiceHigh]

Full answers here:-

RiceHigh's Pentax K100D Full Review

BTW, for the K10D units I tested, there was no major misalignment for the central point, but for the other red points, still there are obvious misalignment (mainly shifted to the left for both the rightest and the leftest point). Plus, there was a strange thing seen is that the leftest point is very dim to be visible, it could be seen only that I moved my eyes to the right when viewing thro the finder. Also, the red point illumination of K10D is obviously not as bright as the K100D (probably for the same LED projector component but the magnification is larger).

Nonetheless, for your case, you can first try to use a needle like object to try to move the position of the focusing screen by pushing its left or right sides depending on the misalignment direction (push in a counter direction). Do note you must do this carefullly and do never touch the surface of the focusing screen (which is plastic and is very easy to be scratched) and you must do this in a clean place to avoid getting dust into the inside of the camera.

My humble advice is do not return your K10D if AF accuracy and actual sensor sensitivity area alignment are okay. Just because it's likely that other K10D will have other misalignments for the red indication (since the design is not perfect, unlike Canon's red indication design, which projected to the craved marks). As it does not actually affect actual performance and is just an indication, you should be able to forget about it. But, it's annoying to see that, I know.

Anyway, if you still find it to be too annoying, try getting another one to see (if you have the luck!).

Honestly speaking, my luckiness record on this issue is as follows (for the central red point):-

1st *ist D: obvious shifted to the left

2nd *ist D (died after 4 month's use): no obvious problem

3rd *ist D: no obvious problem

1st *ist DS (killed by Pentax Japan during "repair"): no problem (perfect here)

2nd *ist DS (an on-loan unit during the above Japan trip): no problem

3rd *ist DS (a replacement unit by local agent because Pentax Japan destroyed the camera): obvious shifted upwards

Current K100D: Shifted to the left (as indicated in my K100D review).

Five other units of K100D checked at Tokyo in August last year: NONE of these are aligning good enough!(?) Two units inspected in HK in August (only one is okay)

Now that the simple maths come, for the non-defective rate on this issue for Pentax DSLRs is:
- Total sample size is 14 units
- Number of acceptable units: 5
- Good rate: 5/14 = 35.7% only!
- Defective rate: 9/14 =64.3%!!

So, you can judge if it is worthy to seek for a replacement! Anyway, good luck!!

Just received K10D a few days ago and noticed the red center AF indicator point is not properly centered in the central viewfinder bracket. Do we accept this as the norm? Is this common? Worth sending it back for a replacement?

Thanks for any feedback.

P.S. Just to clarify, I am referring only to the red, small, indicator box. The real focus point might be perfectly centered within the viewfinder. I have read that it is actually a cross in shape. It is hard for me to verify this. Did some testing by focusing on window pains, etc. but still could not really know for sure.

[Ivan Glisin]

Now that the simple maths come, for the non-defective rate on this issue for Pentax DSLRs is:

I would not call that "a defect" since position of red AF lock indicators has absolutely nothing to do with the position of the actual AF sensors and misaligned AF indicators can not influence AF operation in any way.

In addition, since AF sensor areas are much bigger than AF lock indicators actual position within the AF area does not matter at all and one could not argue "if it is misaligned I can not precisely center AF point on my subject" since you don't know the exact size of AF areas anyway, so talking about any "precision" is pointless. As a reference, center AF sensor is slightly bigger than the center circle used for spot metering.

But I agree with you that Pentax should have paid more attention to better align focusing screen and AF lock indicators. Still, it is really only a cosmetic issue that may be annoying, depending on how much attention one pays to those details. Certainly not a defect.

-= IVAN =-

[Photomy]

Thanks for the info. I really like the handling and design overall. Just getting to know some of the details. I normally only use center focus and then re-compose. Actually, I am using the AF lock button and turned of AF in the shutter. I also just turned off the red lights altogether. Sometime, I may get into the screen and see if it is not set quite right.

Thanks again.

- Tom

[jfdavis58]

Not to pick nits or anything, but are you sure it's the center red indicator that's off?

If I stand just right and look in the mirror for a while, one eye is higher than the other. Heck I bet I got a dreaded 'human defect'. And my warrantee has run out!!! Considering I'm made in the likeness of a perfect model, maybe that model is ....

NAW---OH NO, that would mean maybe everybody is...

Hey, Maybe you got the same problem.

Oh man, I can hardly wait for the Ricehigh exposition and blog write-up on that problem; we all get recalled for a little tweak! Ha!

[RiceHigh]

In addition, since AF sensor areas are much bigger than AF lock indicators actual position within the AF area does not matter at all and one could not argue "if it is misaligned I can not precisely center AF point on my subject" since you don't know the exact size of AF areas anyway, so talking about any "precision" is pointless. As a reference, center AF sensor is slightly bigger than the center circle used for spot metering.

-= IVAN =-

To have precise pinpoint AF, the actual AF sensitivity area should not be made too large. So, the MZ concept is correct here but the *ist D and K layout isn't, regarding this precision.

Indeed, Canon DSLR's AF area is much smaller and pinpointing is much easier.

I actually know about where exactly the AF sensitivity laid of my K100D, but sometimes I found it to be too large and I don't know exactly where it is focused within that large area!

[RiceHigh]

Thanks for the info. I really like the handling and design overall. Just getting to know some of the details. I normally only use center focus and then re-compose. Actually, I am using the AF lock button and turned of AF in the shutter. I also just turned off the red lights altogether. Sometime, I may get into the screen and see if it is not set quite right.

Thanks again.

- Tom

I do the same for turning off the red light. As I do use centre AF point 99% of the time myself, I do not need the flashing red indication anyway and I simply got rid of the annoyance completely, by turning off it. In the past, I mostly switched to the 11-point AF when I gave my DSLR to somebody (usually non DSLR photographers) to take the shots.

Also, I can view better into the focusing screen within the focusing bracket for the object without the interruption of the flashing light, too.

Still, unless Pentax change the design (for two separate sub-systems which requires perfect alignment), it will be very unlikely this problem can be resolved.

[wlachan]

Just received K10D a few days ago and noticed the red center AF indicator point is not properly centered in the central viewfinder bracket. Do we accept this as the norm? Is this common? Worth sending it back for a replacement?

You may shift the focus screen around and see if it matches that indicator better. My DS does. With cheap labour and lower QC than those good old MIJ back in the 80's, some grits are bounded to happen.

[Photomy]

You may shift the focus screen around and see if it matches that indicator better. My DS does. With cheap labour and lower QC than those good old MIJ back in the 80's, some grits are bounded to happen.

Thanks I'll try this - hopefully that simple.

[Harald]

I just tested my K10D yesterday.

Focusing indicators are not completely centered into focusing screen. Centermost one is in the upper part of the circle, not in the middle. This is not an problem as described earlyer in this thread.

But the actual sensor is not in the middle of the circle either. Its center is in the lower part of the circle.

Sensor really is cross shaped and very huge compared to the indicator. Sensor is as big as the center circle itself. And so are the other sersor areas too. Actually there is not space between the sensors in the inner bracketed area at all. Nine middlemost sensors fills the area almoust completely.

Even if you use the center sensor only, you cannot know very precisely where it actually focused because of the large sensor area. I think this is the reason for some miss focusing claims. If there is many details inside this sensor area, it seems to take randomly one of them.

If you want that focus is precisely in the models eye, the eye must be larger in the screen than center circle or sensor can focus to nose instead... :-)

Or you have to use MF.

[Ivan Glisin]

Even if you use the center sensor only, you cannot know very precisely where it actually focused because of the large sensor area. I think this is the reason for some miss focusing claims. If there is many details inside this sensor area, it seems to take randomly one of them.

Very true! My experience teaches me that AF system works with highest reliability only if the subject area covered by the AF sensor is: (1) large and relatively flat area approximately parallel to the sensor plane; (2) represents a pattern of various tones; (3) is not too dark or too bright to avoid AF sensor under/over exposure; (4) is not extremely high contrast, such as pure black on pure white; and (5) does not include bright light sources, or a silhouette against a bright background.

Everything else usually fails and probably is the reason for some miss focusing claims. In fact, popular AF testing chart actually violates requirements 1, 2 and 4 and is NOT a good AF system testing target!

-= IVAN =-

[Harald]

Very true! My experience teaches me that AF system works with highest reliability only if the subject area covered by the AF sensor is: (1) large and relatively flat area approximately parallel to the sensor plane; (2) represents a pattern of various tones; (3) is not too dark or too bright to avoid AF sensor under/over exposure; (4) is not extremely high contrast, such as pure black on pure white; and (5) does not include bright light sources, or a silhouette against a bright background.

Everything else usually fails and probably is the reason for some miss focusing claims. In fact, popular AF testing chart actually violates requirements 1, 2 and 4 and is NOT a good AF system testing target!

-= IVAN =-

Agreed.

In this chart several lines that should reveal the exact point of focus are in the area of focusing sensor.

It is obviously related to lighting conditions, printing quality (evenly printed lines) and other such factors, which line is the real reference within the sensor area and thus focused correctly. There is not any subdivision in the sensor area that could determine the point and so test is not reliable.

Maybe the lines should be more wide apart and the angle the chart is located less than 45 degrees so that only one line is in the focusing sensor area?

[Photomy]

Yes I turned off the red light and tested the actual focus area. Seems to be shifted a bit down from center and does fill almost the whole inner circle but works OK. I remember my Minolta 7000 AF camera had a small horizontal rectangle etched in the screen and nothing else. This was the only focusing area but it was very tight and accurate. You had to tilt the camera sligthly sometimes to help focus on a horizontal line but it worked. I still like the simplicity of the design because I only use center focus 90 percent of the time.