Consistently Underexposing Images, Please Help!

[tom_collins3]

<p>Here are the test images that I took. All available light exposures using spot metering on the center point. </p><div></div>

[tom_collins3]

<p>Test2</p><div></div>

[tom_collins3]

<p>Test3</p><div></div>

[tom_collins3]

<p>When taking all of the above images, with spot metering set, I metered off of the middle AF point and they all showed a correct "0" exposure, but you can tell by the histogram that they're underexposed. </p>

[simon_crofts]

<blockquote>

<p>I metered off of the middle AF point</p>

</blockquote>

<p>What did you point the middle AF point at when you were determining the exposure? And why did you choose whatever it was you chose to point the middle AF point at. That's critical information to understanding what is happening.</p>

[tom_collins3]

<p>On test1, I metered off of the corner of my daughter's right eye (where it meets her nose), on test2 I metered off of the middle of the image (the flower closest to the middle of the image), on test3 I metered off of the middle of the scene on the arm of the chair. I chose the metering points based on where I wanted the focus point, but I realize now that is not the correct way to meter. </p>

[simon_crofts]

<p>So if I understood you right about the bits you metered off, I've cropped into each of those spots in the images and give the histogram for each of them. As you can see, the histogram is pretty much bunched around the middle of the histogram, as they should be. About a third of a stop to the left of centre, which is about what you would expect from a typical in-camera spot meter:</p>

<p><img src="http://www.croftsphoto.com/photonet/Histograms.jpg" alt="" /></p>

<p> </p>

[neill_farmer2]

<p>Tom, It is not unknown for camera meters to be out by a bit, +/- 1/3 of a stop, but a full stop is too much and if this is the case with your camera it needs adjustment. If you dial in a constant 1 stop correction in exposure compensation what happens? Are all the exposures now good or are they inconsistent? I find the exposure meter on my 7D to be different to my previous Canon cameras in that it seems to take a lot of notice of what is under the active focus point (the manual says this is what it is supposed to do). However this leads to your subject being exposed accurately but the overall image can be washed out or dark (or spot on!) depending on what shade was under that AF point.<br>

I echo what others have said, the LCD is not an accurate check on exposure.</p>

[dan_south]

There's a lot of useful information and well-illustrated in this discussion, but it's haphazardly organized. Perhaps

someone could summarize a series of best practices for (a) exposure metering in general (spot metering, reflective

versus incident metering, 18 percent gray, etc.) and (b) digital exposure (histograms, noise, JPEG versus RAW

exposure, blinking highlights).

[simon_crofts]

I was perhaps a bit cryptic last night. What I was trying to say is that, if I have understood the bits that you spot

metered off correctly, then it seems that your spot meter is fairly accurately calibrated. The problem is that you need to

interpret what the spot meter is telling you, and take a decision accordingly how to expose the scene.

 

A spot meter isn't intended to be used like an interpretative meter like a matrix. You don't point it at the scene and set

whatever aperture and shutter speed the spot tells you to. You just use the information given by the spot as a starting

point in order to take decisions about tones around the scene.

 

So, in the picture of the baby's face, the spot meter has given you correct information about how to reproduce the

baby's face as Zone 5, otherwise known as mid-grey. That is what the meter is supposed to do. And that is the result you ended up with. Which is rather a dark result. Caucasian skin should typically be around Zone 4, or +1, or even lighter to look good. So you should have taken the info given by the spot meter, and taken the decision to dial in +1 or +1.5 stops of exposure.

 

Then the exposure would have been spot on.

 

To get an idea about zones, have a look at this page: http://en.wikipedia.org/wiki/Zone_System. Note in particular its

description of Zone 6: "Average caucasian skin". You put the baby's skin as a low Zone 5.

 

That is the problem.

[dan_south]

When I was a novice snap shooter with a point and shoot camera, I used to wonder why my pictures of snow always

turned out gray. "I guess that's what snow looks like to the camera," I mused to myself, cluelessly.

 

 

When I began to study photography seriously, I learned there was some truth to my naive assumptions about exposure. Mid-tone gray is how snow (and everything else) looks to the

camera's internal METER, and to any other reflective light meter for that matter. If you meter royal blue or jade green

or scarlet, the meter will do its job accurately. But if you meter something lighter - white snow, a bright yellow

outfit, sky blue - the camera will get it wrong every time. It will render these things too dark. You, the photographer must

COMPENSATE for the camera's ineptitude. You must make an adjustment and force MORE light in than the camera

things it needs.

 

 

If you point your meter at something dark - black asphalt, navy blue or forest green clothing - you must

compensate by giving the camera LESS lit than it thinks it needs, else all of these objects will be rendered too light.

 

 

Bright objects need more liGHT than the meter asks for. Dark objects need less light than the meter asks for. Mid toned

objects will be rendered more or leas correctly when spot metered. If you can think in these simple terms, improper daylight

exposure will no longer be a frustrating mystery. You'll understand how to compensate (whether manually or with

some built in metering compensation feature) to have objects appear in photographs as you expect them to.

 

 

Strobe (flash) metering and nighttime exposure are other matters entirely and a topic for another discussion.

[simon_crofts]

<blockquote>

<p>Caucasian skin should typically be around Zone 4, or +1, or even lighter</p>

</blockquote>

<p>Sorry, that was a typo. It should have read: "... typically be around Zone 6, or +1, or even lighter"</p>

[picturesque]

<p>I think the confusion results from Tom wanting to know several things at once. First would be whether his meter is working OK. Second would be whether his metering method is OK. Third would be knowing how to quickly and efficiently get good exposures while working a wedding. This results in people telling him to do this and that, working on one of the three issues above.</p>

<p>Seems like his meter is generally working OK. So that's answered. It appears Tom is a bit confused about his metering method--'metering the eye'. It also appears that Tom knows about using a gray card. It isn't clear to me, Tom, that you understand the rough Zone System method that one needs to know in order to implement spot metering as a quick way of determining exposure. The other thing is to know that there is no one 'correct' exposure for any scene, only the one exposure <strong>you</strong> want to show the scene the way you want, given the limitation of the camera's ability to capture the dynamic range of the scene.</p>

<p>As for different methods to use for weddings, that might be a whole different discussion, but you definitely need to understand exposure and metering first. How about you give us some idea where you truly are re understanding exposure and metering?</p>

<p>I've attached a chart that can be used to help you visualize values similar to middle gray and also values that are one stop darker than and one and two stops lighter than middle gray. They are colored because we live in a colored world. So when you are at a wedding, and you have nothing handy to meter, you can pick whatever is around you (that is in the same light as the subject) similar to one of these patches, compensate in your head, and be happy. Of course, this is a very rough method, but it is fast. Note that the green shade in the 0 column is very similar to green grass in the shade.</p>

<p>Note: The image is scanned from a Hasselblad book, so I think the contrast is a bit steep, making the -1 patches darker than they should be. But you get the idea. I'll look up the author and book, to give proper credit.</p><div></div>

[picturesque]

<p>The chart is an illustration in The Hasselblad Manual by Ernst Wildi. I have not found it online anywhere.</p>

[tom_collins3]

<p>Nadine, I have never heard of the zone system before this post so I will need to read up on that! Apparently my understanding of exposure and metering are not what I thought they were! Basically, I understand shutter speed, ISO, and aperture and how each of them affect the image. Based on those understandings, I would set my manual settings and take a test shot. From there I would review my LCD (first mistake) and rarely check the histogram (second mistake). If it looked good on the LCD, I would move on. I have now learned my lesson that I need to review my histogram and not accept what the LCD preview is showing me. I also had an incorrect understanding of using a gray card. I thought that was used solely as an 18% gray reference for fixing color balance if needed in post. I now understand that I should be using my gray card to determine "proper" exposure, or at least a good starting point. </p>

[picturesque]

<p>Tom--that's a fair assessment of your understanding at this point, it seems. I would suggest you do some more research on the zone system and review the exposure book you have.</p>

<p>In the meantime, I'd suggest you start applying the quick and dirty method and see how it goes. That is identifying likely middle gray equivalents in typical scenes using your spot meter, setting the results and seeing how close you are. With some practice, you can get very good at identifying values, which will get you very close with exposures.</p>

<p>Be aware that gray cards meant for white balance are different from gray cards meant for exposure. Be sure you are using the latter, because gray cards for white balance are lighter than middle gray.</p>

[picturesque]

<p>You may also want to read the following.</p>

<p><a href="../wedding-photography-forum/00VOMg">http://www.photo.net/wedding-photography-forum/00VOMg</a></p>

[martin-s]

<blockquote>

<p>"I now understand that I should be using my gray card to determine "proper" exposure, or at least a good starting point."</p>

</blockquote>

<p><br /> Tom, I still think you're complicating things more than necessary.<br /> As the others have explained, given the actual spots you metered, the exposure results are about right. The crucial thing with spot metering is to know exactly what to point it at and then decide to increase/reduce the actual exposure depending on what brightness level you want to achieve for that part of the image. As you found out, light skin needs an increase in exposure.<br /><br />In the old slide film days you could for instance point the spot meter at a person wearing a white (well-lit) shirt, measure just the fabric and then increase the exposure by 2.5 to 3 stops. That would reproduce the white fabric faithfully as white while still preserving visible detail, hence not blow it out.<br />Spot metering almost always requires manual intervention and is therefore best used in settings, where you have enough time to carefully set things up, studio-, architectural- or landscape photography.<br /><br />So my advice, at least for the time being, would be to switch your camera to evaluative- or matrix metering and bracket your images - 2/3rds of a stop is a good starting value.<br />Well-known street photographer Jay Maisel said in a seminar "Bracket your images, it's much quicker and easier to choose the best exposure than to spent time fixing an image later."</p>

[june_salomon]

<p>I have a 7 D as well and experience similar issues. It may have to do with the built in metering system, which I have yet to be able to over ride. I have used grey cards and custom white balance. Very frustrating even in M mode</p>