[explr1]

I've had my K10d for a few months now and am getting tired off it taking under exposed crappy pictures that have become wasted memories. Why can't the K10d meter correctly when Canons and Nikons can? I took over 400 pictures on my trip to Mystic and only had 50 or so keepers. I've come to the conclusion that the
K10d sucks!!! I'm sure there will be people that are going to blame the bad pictures on me, but my wifes lowly 3.2 mega pixel Canon PowerShot A75 took better pictures of the same objects and exposed them correctly.

Some of the under exposed pic's from the K10d.

[serendipity]

I'm no expert, but the shutter speeds (in the exif data) looks too fast and maybe a stop or 2 down (higher f number, f8 at least) would have produced lighter shots.
Someone with more expertise than me will put you right But it's NOT the camera, I'm afraid it's the operator


Mick

[MrApollinax]

I've had this happen a few time to me before. Turns out I was metering off of a really bright portion of the frame. It was explained to me that in the mode I was in the camera would attempt to expose the frame to not blow out the highlights automatically resulting in a darker frame. His advice to me was to shoot more in the semi-manual/manual modes to understand how the settings effect exposure and then take more pictures metering off of different levels of brightness. Even though i have an older K100D the above exercises have helped me greatly.

[fontku]

I agree. Pentax metering usualy underexposes this type of pictures with higth bright portions of the frame (more when more sky is present). You can compenstate it manualy (+ 1 EV aprox ) keeping the histogram centered, or shot raw and post-process.

[RBellavance]

You will notice that your camera managed to keep a lot of detail in the sky (especially striking in the first shot). I doubt your wife's A75 did the same.

Pentax's multi-segment metering is traditionally bias towards not blowing highlights. Once you know that, you have several choices:

1. Use positive exposure compensation in situations where there are areas of the image significantly brighter than the others (esp. the subject).

2. Use center-weighted or spot-metering to meter for your subject

I would also recommend looking at the image review after you take it, and turn on the histogram display on it.

The K10D, like other DSLRs, has the potential to take much better pictures than a P&S. But it requires more care and thought in its use to do so.

[Nesster]

You can usually recover most of these via curves etc, but it is a royal pain. If it is any consolation, the K100D also does this. From the convenience pov, Nikon and Canon do have it over Pentax here... though with their systems that dread highlight blowing is more likely.

[RBellavance]

Hmm, I just noticed you were using spot metering. That it, IMHO, a very difficult metering mode to master, especially if you focus-recompose and don't use EV-lock.

May I suggest trying multi-segment metering for a change ?

[WMBP]

I've had my K10d for a few months now and am getting tired off it taking under exposed crappy pictures that have become wasted memories. Why can't the K10d meter correctly when Canons and Nikons can? I took over 400 pictures on my trip to Mystic and only had 50 or so keepers. I've come to the conclusion that the K10d sucks!!! I'm sure there will be people that are going to blame the bad pictures on me, but my wifes lowly 3.2 mega pixel Canon PowerShot A75 took better pictures of the same objects and exposed them correctly.

I get the impression that you don't think it's your fault. So you have "come to the conclusion" that the K10D sucks. But I'm afraid that you've come to the ONE conclusion which is clearly wrong. Did you purchase the K10D without being aware that many other photographers using the same tool have taken really great photos, that the camera got terrific reviews, etc.?

The alternatives you should be considering are: (1) you didn't use the camera properly, or (2) your particular camera is defective. It could be (2), but (1) is MUCH more likely.

I looked at the EXIF info for two of the photos (the first and the last) and it looks like you were shooting on Auto mode, and that in the first photo you were shooting with center spot metering, while with the other photo you'd switched to center weighted metering.

Now, notice that in the center of the first photo there's a big white sail. Notice also that it's pretty well exposed - it's the rest of the picture that's got a problem. If I've got these details right, then it looks to me like the camera did its job just fine on that shot.

About the last photo (with the woman stepping off the ship) I'm a little less sure what to say. Looks like center-weighted metering here picked up the bright back lighting above and to the right of the woman and that had an effect. Not sure but that's a possibility.

*

Anyway, a couple suggestions.

1. Take the camera off Auto. I suggest you switch to P, learn to use the green button.If you want a point and shoot, borrow your wife's.

2. Configure the camera so that it displays both the histogram and blown highlights in the display screen after you take a shot, and take a quick look at that feedback every time or as often as you can. These can be tremendously helpful. Don't pay so much attention to how the shot LOOKS. The screen is pretty small and it's actually hard to tell. But you can learn a LOT from the histogram and the highlights flasher.

3. These shots were rather challenging exposures: a big dark thing in the foreground, with bright sky in the background. You need to understand that, when the dynamic range of the photo is quite large, as in these photos, you've got a real problem to deal with. If you want to get the most out of your investment, learn a bit about exposure. Bryan Peterson's book Exposure is a great start, I also like Chris Weston's books on the subject.

4. In this case, you have two easy choices. First choice: Set camera to spot or center-weighted metering. Point camera right at an important but dark area of the ship and hit exposure lock button. Recompose the shot quickly and shoot. You risk blowing some highlights in the sky here, but that may be inevitable. And the ship should be nicely exposed. Second choice: Put camera on matrix metering, take a shot, look at histogram and flashing blown highlights, and adjust the meter as necessary (+/- button) and shoot again. What's the difference between these two? If you meter on the dark areas yourself, you are making the decision about what's important. If you use matrix metering, you let the camera simply average everything. I don't use matrix metering a lot but when I do use it, I get results that are actually better than I would expect. So I think it's worth a try.

5. There's a third, less easy choice, and it's the one I'd probably use. I'd probably meter on an important dark area of the subject, but then deliberately adjust the exposure down a little (-1/3EV). The goal here is to get the shadowy areas pretty close to properly exposed, but at the same time to avoid blowing the sky. I would then bring up the shadowy areas in post-processing.

6. Finally, consider saving your raw files. They will give you more data to work with in post processing and this can be really important.

Will

[egordon99]

LOL, I was in Mystic last summer and accidentally switched my K10D to Spot Metering (since it was on a knob and not buried in a menu like my DS) and I got a bunch of dark shots until I realized what was going on....


Anyway, read this thread - Anyone else have issues with blown highlights? [Page 1]: Canon EOS 1000D / 450D - 300D Forum: Digital Photography Review

I bet your wife's P&S shots had completely blown out skies.

Here are some of my Mystic shots -

[AndrewG NY]

I don't think this is completely K10D's fault--obviously others make it work.

In this case, for the first two you were using spot metering--matrix probably would have been much closer to what you want. Spot metering completely puts the burden on you to determine proper exposure. If you put that spot (a very small region of the image, right in the center) on 18% gray, you're golden. If you put it on white, you're going to underexpose your shot, as the camera will want to treat that spot as 18% gray--this appears to be what you did in the first shot--the spot is on the white sail, so the camera tries to make it gray. This is what pretty much all camera meters are trying to do--adjust exposure so that the region the meter is concern with averages out to 18% gray. The second is probably similar--the spot meter probalby caught a piece of bright sky through the ship rigging.

The third one doesn't look all that bad considering the amount of sky in the shot. I would have expected that to possibly require some positive EV comp for a better exposure. Fourth as well--backlit subject, probably should have added positive EV comp or fill flash, though subject might be a little far for fill flash.

You also used center-weighted metering in shots 3 and four--matrix probably would have reduced if not completely eliminated the need for EV comp. Center-weighted generally doesn't adequately compensate for backlit subjects or light subjects on dark backgrounds.

As far as comparing to the P&S--had you left the K10D on matrix it probably would have done a much more comparable job but it may be a bit more conservative, trying to avoid blown highlights. Some point & shoots I've used seem a bit more willing to allow highlights to be blown, probably under the assumption that their operators won't be real willing to do much post-processing.

[AndrewG NY]

I agree. Pentax metering usualy underexposes this type of pictures with higth bright portions of the frame (more when more sky is present). You can compenstate it manualy (+ 1 EV aprox ) keeping the histogram centered, or shot raw and post-process.

While it's true that RAW will improve the ability to correct this during PP, you'll still get better images if you do your best in-camera. Bringing up the shadows will also increase noise, and by underexposing you're generally not collecting as much detail and texture as you might have had the image been properly exposed in the first place.

This has always been true with any camera. Meters are not infallible, it's the operators responsibility to know how they work if they want the best results.

[AndrewG NY]

A few other responders noted that it is probable that the wife's P&S blew out the skies. While this may be true, apparently this might have been acceptable to the original poster--he was essentially complaining that the subjects were too dark. I would tend to agree though that the ideal result would be maintaining some amount of sky detail if it represents a substantial part of the frame. Balancing these is the trick, and DSLR such as K10D has the tools necessary to do it with a good degree of control, much more so than that P&S.

Another thing to keep in mind is that when shooting JPEG (as no doubt the P&S was doing as well) the contrast setting can have some affect on what range the image will show and if contrast is set high, you'll lose even more of the range that is already difficult to capture--this is another reason that RAW is superior as you get to make that call later without image quality penalty. If shooting JPEG though, it may be a good idea to keep contrast low in shots like this--it's more effective to add contrast during post-processing than it is to try to reduce it.

[WMBP]

Here's a shot I took in NYC last June.

New York City, near the Fulton St market

It's not a very good photo, in fact, it has lots of problems. I offer it not because it's proves how easy it is to do this right if you know what you're doing, but almost to make the opposite point - that a hard exposure is just hard. I didn't do a very good job here, but I think I know what my mistakes were.

I was facing the same problem the OP was facing: bright sky, but a very dark area of the photo that was really important. I keep saying "significant" and "important" because some areas of the photo are NOT very important and you don't worry too much about them.

Shot was taken in manual mode. I wanted a fast-ish shutter to freeze the motion of the people, the flags on the ship, etc. I settled on 1/250th - just a guess from experience. I could have gone a little slower but not much. I also wanted a lot of depth of field field so I stopped down to f/14. This was a mistake: f/11 would have been just about as good and f/9 probably would have been fine, too.

The shot that I ended up with was pretty underexposed - not just the ship, but the sky as well. The histogram was bunched right in the center and was nowhere near having blown highlights. My bad. I was wandering about the city with my family and was not taking the time to review my shots and reshoot. If I had taken one more minute here, I would have fixed the exposure (1/250th sec, f/9) and made a better shot.

Fortunately I was saving raw here, so I had more data to work with in post-processing than I would have with a jpeg.

Will

[omega leader]

The best thing I did was change the setting that tied exposure to selected autofocus point. I then use center-weighted and get much better results that other modes, as normally I want what I'm focusing on the be correctly exposed.

Try it.

[augustmoon]

I went through the same type of experiences when I switched from Canon to Pentax. The Pentax metering is different and takes a while to get used to, especially if you recently switched from Canon. I would say, that Pentax metering is closer to how I used to meter for eons when I had older film bodies and medium and large format cameras.

I would also say that the Pentax metering is both more accurate and requires more input from the user. Canon built theirs, and from what I remember Nikon as well, to expose as a P&S. They would prevent underexposure at the expense of blown highlights. After I switched from Nikon to Canon, I still had some pro friends that still shot Nikon, but all and always in Manual mode.

Pentax metering is all working second nature to me again, just like the old days. The old days in 35mm film for me were with Pentax, and I remember that being a main reason why I liked Pentax so much was their metering.

Personally for my habits, I prefer using for the bulk of my shooting, the Center-Weighted Average.

It's the metering technologies from Canon and Nikon that are less "traditional" I realize that I had let Canon figure out and adjust for things that I should have been in control of and not the camera. I'm not saying that the Canon metering was superior, it was quicker to just point and get a keepable snap than with Pentax.

But my Pentax shots are actually more properly exposed and yield better images. In re-looking over my Canon image catalog out of Lightroom. I can see where a lot of these keepable snaps from my Canons have blown out highlights and highlight edge fringing and hot spots and all kinds of issues that I wasn't paying attention to. In tricky contrasty lighting, the Canon will over-expose to give a result, not unlike what people are used to with the P&S. The Pentax will expose for the brighter area, preserving it, at a cost of underexposure of the foreground or subject (and resulting noise when you take the levels up in PP)

My early "mistakes" from the Pentax K20D when I first got it, where underexposed shots like yours, where the meter was "wrong". The meter was wrong and the camera screwed up the exposure and the final image. I started reviewing each shot's histogram after taking some pics, then all my old training started to trickle back into my brain.

Also I have lenses for my Pentax that spread from 1980 to last week, and some of them meter a little different than each other, so I've gotten to know how each of these lenses react in low-light incandescent, outdoors, studio flash etc... So I keep the ±EV in general dialed in when I put each of my lenses on.

It's really up to me (not the camera) to select (in a tough shot where the range of the scene falls outside of the range of the film or digital sensor) what I want to be exposed and what I'm willing to let fall off into unrecoverable black, or clipped out to white. Or whether I'm wanting to try to go for the middle or take one of each for a future HDR attempt.

I know (like I learned 30 years ago) that when I am metering against a bright sky, I'm gonna need an extra stop or so above what the meter tells me, so I dial in some additional +EV Comp, or for really tricky lighting (and another reason I love these Pentax DSLRs) I'll switch to Manual mode, meter on what I want to be correctly exposed, then shoot the entire scene I want, then look at the histogram and see if I can get it better or not, or maybe say forget it and find a different angle or a different view or crop. When I'm shooting a skier against a snow slope, I know it's gonna be trash unless I switch to Manual and open up at least 2 stops or more. Same for surfers on the waves at sunset (Pacific coast)

I can see how I got lazy with my Canon 20 and 40D and was basically using it as an expensive interchangeable lens DSLR "point and shoot"

I think that you have a great camera with a great meter, and if you wanted to, you could go out and play and have fun and do some tests and get a feel for it. If all this sounds awful and unappealing to you, perhaps you'd have more enjoyment with a P&S and forget about the effort and just enjoy your trips without having to worry about anything.