[jjfdvm72]

I have read previously that the K10D tends to slightly underexpose the pictures. Yet it seems that with my K10, its not a consistent, mild underexposure, but rather an occasional WAY underexpose pics. From what I gather it is consistent with P, TAv, Apperture priority and Shutter priority. I have tried all 3 exposure modes and seen no difference. Could something be wrong with my light meter? Here are a couple of samples.

Hope the attached pics come through...

Attached Images

   

[bdann]

Are you using the kit lens?
Many of my photos using the kit lens are underexposed, I have not had the same issue with my older lenses.

[Peter Zack]

I haven't had the problem myself with the exception of a couple of K mount lenses that needed a little Ev comp help and I assume it's due to the crippled mount. Some others have found the 16-45mm f4 have shown a minus 1/2-1 Ev underexposure. but with all the glass I've used other than the above, I've not seen this on 2 different bodies.

How is the camera outside in daylight and what lens are you using in the examples?

[davef]

I believe the exposure calculations will be thrown off by the backlighting in the first image and by the bright object in the foreground in the second image. Try spot metering on the actual subject, which I assume is the people. You may have to use manual exposure with this reading if the main subject is not dead center in the image. This will result in the highlights being "blown out" or overexposed but your people will be fine. The other options are to use fill flash to adjust exposure on the subjects or to shoot raw and post process to "rescue" the shot. BTW, I expect any metering system (average, center weighted or center spot) would struggle with these examples. Dave

[MikeH]

The flash metering has been fooled by the reflective foil packaging on the table.

Did you review the photos in camera after shooting? I'd have either moved the packaging or re-shot with exposure compensation.

[jjfdvm72]

Are you using the kit lens?
Many of my photos using the kit lens are underexposed, I have not had the same issue with my older lenses.

I am using the kit lens

[jjfdvm72]

I haven't had the problem myself with the exception of a couple of K mount lenses that needed a little Ev comp help and I assume it's due to the crippled mount. Some others have found the 16-45mm f4 have shown a minus 1/2-1 Ev underexposure. but with all the glass I've used other than the above, I've not seen this on 2 different bodies.

How is the camera outside in daylight and what lens are you using in the examples?

The camera is definitely worse with the underexposure inside than in the daylight. It almost seems it takes the brightest point and adjust it to make it a middle tone. Therefore, having the midtones seem like dark tones. Does that make sense? Its almost like it shifts the whole histogram to the right by making the brightest spot the mid point on the histogram

[jjfdvm72]

I believe the exposure calculations will be thrown off by the backlighting in the first image and by the bright object in the foreground in the second image. Try spot metering on the actual subject, which I assume is the people. You may have to use manual exposure with this reading if the main subject is not dead center in the image. This will result in the highlights being "blown out" or overexposed but your people will be fine. The other options are to use fill flash to adjust exposure on the subjects or to shoot raw and post process to "rescue" the shot. BTW, I expect any metering system (average, center weighted or center spot) would struggle with these examples. Dave

I understand that it would read the foil as being bright, but wouldn't it still "average" it out to the midtones?

[mattdm]

I understand that it would read the foil as being bright, but wouldn't it still "average" it out to the midtones?

Which then makes the real midtones come out too dark.

[jjfdvm72]

would there be a way to "test" the light meter? It just seems that my small canon a70 point and shoot digital does a LOT better job measuring the right exposure. Don't take me wrong, i love the k10, but I feel I'm missing something out with the exposures

[PentaxPoke]

I tried to look at the EXIF data from your pics, but it is mostly gone. Did these come directly from the camera, or did you edit and save the file in some way?

I was curious what type of metering your k10d was set to.

[jgredline]

I had the exact same problem...I blamed everything except the photographer (me)...Turns out the problem with mine was the Photographer...

Try this.Put the camera on TAV then hit the little Green button, If the flash thing blinks pop up the flash and take a look at the histogram after you take the pic...This should give you a good indication of hoe the light meter is working.

[jjfdvm72]

I tried to look at the EXIF data from your pics, but it is mostly gone. Did these come directly from the camera, or did you edit and save the file in some way?

I was curious what type of metering your k10d was set to.

I actually downloaded them to photoshop elements and then exported them to my desktop. I am about 99% it was set to Multi-Segment. I was told by a photo sales person that I should always set it to Multi-Segment. Is that so?

[jjfdvm72]

I had the exact same problem...I blamed everything except the photographer (me)...Turns out the problem with mine was the Photographer...

Try this.Put the camera on TAV then hit the little Green button, If the flash thing blinks pop up the flash and take a look at the histogram after you take the pic...This should give you a good indication of hoe the light meter is working.

I am asuming I would be looking for a "bell shape" histogram? Does it matter what I am shooting a picture off?

[PentaxPoke]

Something just doesn't seem right. The picture on the right is from the camera held vertically? There is no background light in that case, there are plenty of mid tones, and as long as you weren't on spot metering, that jiuce box shouldn't have thrown off the multi-segment metering that badly.

Something is set that we are not thinking about. Were any pictures from this event fine, or were they all this badly underexposed? No one can tell me that this is typical of the k10d.

One more thought...Did you check what the exposure values were for these pictures? With the original jpeg we could see what all the values were in the EXIF (fstop, aperture, ISO). I just realized you used the kit lens and were inside, and did not use a flash. Is it entirely possible that this is just an underexposed pic beyond the limits of the camera with that lens? Was either the aperture or shutter speed values flashing when you metered? When I am indoors I don't even bother taking out my zoom. I use my 50/1.4.

Oh yeah, I thought of something else...what is your auto ISO limit? You may have that set so that your camera could not auto adjust the iso high enough for an exposure indoors. The default is something like 400.

All questions we could answer if we could see your original EXIF... Can you post the original jpeg (not saved out of Photoshop) of the right side picture? I really think we can solve this if we know the exposure settings and if you can tell us the general lighting of the room (dimly lit, windows only, etc.). You are absolutely right in expecting that your $800 k10d should have nailed that picture when just about any P&S could. (Heck the reason many of us have DSLR's is to get the best possible pics of our kids!) It's just that the k10d has some default settings that you may need to change for the way you shoot. The P&S no doubt would have fired the flash, but with the proper settings, the k10d may have been able to capture a much warmer shot without flash.