Hey ya'll!

It's me again

Ok so most of you know that I shoot with a 40D, I couldn't afford the 5D when I first started out. I love the 40D, and have done all kinds of monitor calibration and that sort of technical stuff.

BUT...

I think this might be a camera thing. I shoot in manual and I have practiced and practiced NAILING my exposures. I would say most of the time I do nail my exposures, but my images are still coming out dark, even kind of gray most of the time, and they look the same in Lightroom when I edit them. I end up having to increase the exposure in Lightroom a LOT, and by the time I get done with them, they would be "overexposed" if I had shot them that way, in camera, by most standards. I have to increase the exposure first before I can do any other adjustments. Oh, and I shoot in RAW.

Is this normal? Do most of you have to increase your exposure in PP? Or is this where the 5D excels compared to the 40D? Does anyone else have this problem? PLEASE tell me it's not just me...

HELP!

Love ya'll!
L.

Can you show some of those images? How does you histogram look like?

Yes please. Post a photo..and the matching histogram, so we can see.

Hey Ya'll! I can when I get back to my office, sometime this evening. I can post a photo, but how do I post the histogram? With a saved screenshot?

Define "nailing" the exposure. My guess is you're taking images of very bright objects (like a sky or a white dress) and lining up the exposure meter in the camera right in the middle. That's guaranteed to produce an underexposed image.

See, there are two kinds of light meters. There are incident light meters, like a hand-held light meter, that measure the amount of light falling on the meter itself. This type of meter is always going to be accurate. It is literally measuring the light falling on the subject. Of course that means it has to be held at the subject's location. Difficult to do if the subject is far away or unreachable. Then there are reflective light meters, like the one in your camera, that meter the light reflected by the subject they're pointing at. The problem with a reflective light meter is that different subjects reflect different amounts of light. A black object is going to reflect less light than a white object.

So what can you possibly tell from a reflective light meter reading? Obviously, the designers realized this problem, so they calibrated reflective meters to give you a correct reading, assuming the subject is medium (18%) gray. This is a fair assumption for most scenes. Say, a bride and groom in a park. You've got a white dress, a black tux, medium skin tones, dark grass blue sky...the whole thing averages out to gray. You'll probably get a decent exposure by putting the meter arrow right in the middle of the range. Now zoom in on the white dress, though, and the assumption that the subject is 18% is no longer valid. Instead, the meter will tell you to underexpose the image, "forcing" the white dress to be...18% gray. You'll be off by 1-2 stops in this case. The opposite is true if you zoom in on the black tux. It'll assume that black is gray, and the "correct" meter reading (where the arrow is in the middle of the range) will, in fact, overexpose the image by about 2 stops.

Try it for yourself. Go into your closet and get a white shirt, a black shirt, and a gray shirt. Lay them out somewhere with bright, even light. First point your camera at the gray shirt (close up, zoomed in so the gray color fills the frame), and look at the meter reading. Now point it (close up, zoomed in again) at the white shirt. Watch the needle now jumps 1-2 stops "overexposed." Now to the black shirt, and watch the needle jump 1-2 stops "underexposed." But how can this be? After all, the light is even...it's the same light falling on all three objects...and yet there's 3 different exposure readings!

It's because the assumption that the subject was 18% gray was only valid for the one gray shirt. That assumption was off for the black and white shirts, so you, as the meter operator and interpreter, need to take that into account when reading the meter and setting your exposure.

There's nothing wrong with your camera, you just have to interpret the instrument. It would be like if I, Matt Radlinski, photographer, looked at somebody's cat scan and couldn't tell whether or not they had a brain tumor. "This cat scan machine is broken! It won't tell me if this person has a brain tumor!" Um, no. The machine works fine, I just don't know how to interpret the results

So, your job as the photographer is the look at the scene, evaluate the "grayness" of the scene in your head to figure out how far off the meter's assumption is, and compensate accordingly. Or use an incident meter

Does that help?

Yes, I think that might be it. I posted something last week about having trouble not knowing what to expose for. I have read so many books and find that a lot of times they seem to be exposing for the sky, in order to get the deep blues. But how could that not cause the rest of the scene to be underexposed? Sorry, I just need help!

I think you're missing something simple not something difficult.

When you are importing, Lightroom isn't accepting the camera presets.

So your RAW file is coming out unprocessed. The camera takes your "grey and dark" RAW and says "with this particular camera preset it looks like this."

Sometimes you can import into LR with *nothing* selected and you get a flat, crappy looking shot SOOC.

... I don't use RAW so I don't know the right words to use here, but I'm 95% sure this is your issue.

Hey Matt! I don't use any camera presets, if i'm understanding your comment correctly? Also, can you send me the link to your tutorial for putting your logo on your images that are posted to the blog?

When you shoot RAW your display on the LCD is not showing a RAW photo - it's showing a RAW with a specific conversion applied to it. As far as I understand it, the LCD isn't showing you the RAW image. The RAW image is "unrefined" so it's neither contrasty nor sharp. You have to apply contrast & sharpening when you go from RAW to JPG - during that conversion. When you convert, though, you need to pick how you want the image to look. If you're importing with "nothing" in LR, you're importing the "real" RAW file, which i think would be dark, grey, low contrast, and unsharp.

As far as the tutorials, I removed them from my site. The best you can do is search http://www.cutframe.tv and see if that one is ther.e

Yes, the description you posted above is characteristic of my images SOOC. so is there a way to fix that?

Create a preset that makes your images look fairly close to "perfectly exposed" and use that on import - you can import with a preset. Just set that to "RAW IMPORT" or something, then everytime you import, use that.

Ok, thanks Matt and everyone that helped! I really appreciate what you said up there Matt R! I will try to take that into account when I am shooting as well. OSP is seriously the best

Oh and Matt A. Don't forget to send me the link to your tutorial on how to post your logo to blog images, the one you create an action for!!

Sure, if you want those deep blues, you need to expose for the sky. And if your subjects are not being lit by light of the same intensity (ie, the bare sun), yes, they're going to be underexposed and in shadow. Now what you're running into is the limitations of the exposure latitude of the sensor (or film). The human eye can see about 20 stops of light. That means even in bright noon-day sun, I can still make out detail in objects under a shaded rock. Your camera, on the other hand, can only make out about 4 stops of light. So if I expose for the sky and the open areas lit by the sun, the part of the scene under that rock shade is going to be perfectly, utterly black. And if I expose for the shade under the rock, the stuff lit by the sun is going to be pure white. This is why in order to take pictures outside in daylight you need to be adding light to or subtracting light from the scene in order to get a pleasing balance between the subject and the background.

For instance, this image was exposed for the sky, late in the day as the sun was setting behind the couple. There is far, far less light falling on them from the front than there is illuminating the sky behind them. Hence, they're dark. I'd have preferred a true silhouette, but, what can you do...those are the limitations of the light available at that time of day. There's probably 3 stops more light in the sky behind them than is falling on them from the front.



Now this image has the same exposure, except I added a light off from the side, set to drop as much light onto the bride as there is light in the sky behind her. Now it's balanced, because I exposed for the sky (after all, I can't change the amount of light in the sky ) and then added light to the scene to get a pleasing balance between subject and background.



Your job as photographer is that of "light editor." Your eye sees 20 stops....your camera sees 4. You have to look at a scene and be able to "edit out" 16 of those possible stops in a scene, so you know how the image will look on film before you take it, set your camera to properly expose for the things you can't control (after interpreting the reflective meter data), and then adding light to or subtracting light from other parts of the scene over which you may have control.

Does this help?

YES! Thanks Matt! You're wonderful! I think this might be the easiest explanation of exposure EVER! I have read and read and practiced but still couldn't seem to figure it out and what I should be exposing for and what I shouldn't.

You're welcome Lauren

Hey Matt! I just sent you a PM on more about this. I really appreciate yours and everyones help!

So given the example of the shade under the rock and the bright sky.

If I wanted to expose for the detail under the rock, I would need to meter for the light under the rock, and in order to not have the sky totally blown out, I would need to essentially underexpose the image according to the meter reading for the rock in order to capture the blueness of the sky,

OR add flash somehow to the shadows in the rock and meter for the sky if I wanted to blueness of the sky but still wanted to capture the details of the rock?

That's half right. Yes, if you want the detail under the rock, you'd have to expose for it. However, by doing so, the sky will be completely blown out.

Let's say we're in noon-day sun. What's your exposure? Sunny 16 rule, natch, so we're at ISO 100, 1/100th of a second, f/16. That's the exposure for the open field and the blue sky. What's the exposure under that rock? At ISO 100, 1/100th of a second, it's probably f/2.8 or f/4.0. If I set my camera to f/2.8 to get the detail under the rock, I'm now overexposing the sky by five full stops. And your sensor only has about 4 stops latitude (above and below the proper exposure), so the sky is now three stop beyond the possible farthest limit of what your sensor could handle. Good-bye blue skies, that sky is now white.

On the other hand, if I expose for the sky by setting my camera to f/16, I'm now underexposing the stuff under the rock by 5 stops, or 3 stops beyond the possible range of the sensor, so it's pitch black there. There is NO single camera setting that can properly expose both the sky and the shade at the same time without modification to the light in the scene itself.

Remember, the "artist's brush" isn't the camera...it's the light. The camera's just the canvas.



Now you're talking. If instead you set your camera to expose for the sky, f/16, and then added a light set to about f/16 to the shadowed area under the rock, you'd be able to expose them both at the same time. We can't subtract light from the sky, but can add light under the rock.

This is why photographers who don't want to work very hard *cough*lazy*cough* stick to cloudy days and open shade. The backgrounds and the subjects will all have fairly even light. It's easy, but it's flat, boring and lacks depth. Give me some bright, directional, even harsh light any day, so I can use that to create something far more dramatic, three-dimensional and beautiful.

Matt...you're such a show off.


Extremely helpful explanation. Thanks.

Matt when is your book coming out?

Well, it's sooo easy to accidentally push the EV compensation button on the 40D (the 5D is no exception).
Make sure that you are not compensating, in this case under compensating your EV settings.


Good luck.

Matt should write "Understanding Exposure, for the rest of us" book

looooooooove it!
OMG the light! the colors!! = amazing!
geez look at the dimensionality to that dress! wow SO IMPRESSIVE

you can't post that image and not tell more...

shesh that must have been a heavy-duty super-duper kind of powerful light...?

will you share settings, you know, getting us onto Chapter 2 of your Exposure for the Rest of Us book-in-the-works....

[note to self: no sharing of my off-camera lighting attempts; clearly i haven't gotten it just quite yet....]

Wow, thanks Amy. Making me blush...

No, it wasn't anything fancy...just a Lumedyne light with a 5" parabolic and diffusor off to the right of frame.

Protip: If you ever want to know the camera settings used for most any image posted online, if you open it up in Photoshop and select File->File Info... and then click "Camera Data" you can see the EXIF data that will show you what the camera was set to.

That doesn't necessarily tell you anything about the lighting, natch. Like I said, camera = canvas, light = brush

So this image was on a Canon 1DmkIII, probably with the 24-70, zoomed to 46mm, ISO 800 1/60th of a second at f/6.3. All those settings were determined by using the reflective metering in the camera to meter for the sky. The sky's a touch under just to get the cloud coloring nice and dense and rich. Then it's just a matter of adding an off-camera main flash (simple Lumedyne light...love those things. Powerful, portable, and practically indestructible) to expose her at f/6.3, also. I would have used a lightmeter (Sekonic with the built-in PocketWizard transmitter) to determine that, but after using the same Lunedyne strobes for the last six years, I can pretty-well eyeball it now

It's that cross-lighting that makes the texture in the dress really stand out. That's why direct flash and flat light (like cloudy days and shaded areas) produce such flat images...with no cross-lighting for shadows, there's no dimensionality to the light. It's highlights and shadows that give the illusion of three-dimensionality on two-dimensional images.

One thing to note, though...the lighting on this image is not perfect. I mean, it's beautiful and lovely for a quick job at a wedding, but notice the color of her skin in the shadows. Since I had no fill-light (the fill was just whatever the ambient light falling on her was. Probably 2-3 stops under the sky and the Lumedyne), the color temperature of the light creating the shadows (probably ~2800K) is far, far lower than the color temperature of the flash (5500K) for which the image is balanced. Because of that, her skin in shadow is very red-orange. If I had more time (and another light) I'd have added a fill flash, also, to even that out.

Hope that helps

Matt

P.S. I'd love to write a book, but I don't have time! I've been wanting to host a lighting workshop for years, but, man....time!

Matt! Thank you SO much for all the tips, it was very helpful and I think it's starting to come together for me! I really do appreciate everyone's comments and advice to this post

L.

My tip in mad wedding situations where you need it fast: Spot meter off the face and push it (increase exposure by) 1 and a half stops to get it as close as damned

Have fun, exposure is a nightmare until that day when it all comes together, clicks and becomes instinctual That day, you remember forever

Matt is an open book.

Here's an example of shooting in full, although patchy sun. This is about 4:30 in the afternoon, with the sun blasting down on this area. Now we all know what happens to blonde hair and white shirts in the sun right? Ka-Boom! They explode with highlights. So take what Matt said about the Sunny 16 rule and start there. This wasn't a complete noonday hot sun thing, but it did have some issues to overcome with managing the highlights, while keeping shadow detail. I didn't want to use f16. I wanted to use a shallower DOF. I wanted f4, but couldn't get there with the sync limitations of the SB800 I was using in the 40" umbrella. I was limited to 1/250. So that put me at f5. The point illustrated is that what Matt was saying about lighting the shadow to balance the highlight. I couldn't turn down the sun, but I could use light to balance it.

Thank you guys! This was extremely helpful to just read!!!!

I really love this shot. Very Gone with the Wind look to it. I prefer it to a complete silhouette. I think it gives the picture more interest.

Matt... awesome advice!!!!! Doesn't it feel good to be NIIIIIICE?

Matt, of course, is 100 percent correct. Photography is not about what the meter is telling you. It's about exposing for the situation and being able to adapt. I cannot imagine any reason for any professional photographer not be be exposing manually in this day and age. You have instant feedback on your LCD. This is the true power of digital. Give your clients options. Expose for highlights... expose for skin. Make that camera your bitch... (Sorry, someone had to say it).

God, I love OSP. I feel all funny inside.

This is interesting because I constantly have a hard time with white shirts on a bright day. I just hate them.

I know. I always ask them to not wear the white for the outdoor stuff if it's sunny, but do they listen?

You guys are wonderful! I had 2 shoots this weekend, one last night, and then one this morning early on the beach... and they ALL...wore....white!!!!!!!!! So we'll see how it comes out, I took y'alls advice and I hope they came out better. I even used some diffused flash on the one shoot this morning

Post is coming soon

Hey thanks Shane. To a point though, photography (in the small technical area of exposure) is about what your meter is telling you...but you have to interpret what the meter is telling you. If all you're doing is lining up that indicator with the dead-center of the exposure meter in the viewfinder, that's not going to give you a good exposure unless your subject is, in fact, 18% gray. So getting good exposures from an in-camera meter is about looking at the scene, evaluating it's "grayness" and then setting the camera to give you a good placement of the exposure indicator for a scene of that grayness. So, for instance, in the case of doing a close-up detail of a bride's white dress, that probably means you want the exposure indicator to read 1.5-2 stops over the middle of the exposure meter.

And like you said, the beauty of digital is that, immediately afterwards, you can check your histogram to see how close your guess was. Histograms are great tools, but, just like the reflective meter, they're tools that have to be interpreted by the trained and experienced eye. There's no such thing as a "correct histogram." There's only a "correct histogram for a particular scene." Once you're able to understand the relationship between what a scene looks like, and what kind of histogram a scene of this type should produce, then you're able to use your histogram to check your exposures.

Also, I received a PM from someone with a further question about this topic. Since I went to the trouble of writing out a reply, I figured I would post it here for everyone to see. I'm sure if one person has a question, others have the same one.



Your flash output is, for lack of a better word "power." Now, how that flash power turns into a particular f/stop depends on 3 things: the power of the flash, the distance from the subject, and the ISO you're using for the image. For a given flash output power, the higher the ISO, the greater the f/stop of light that power will produce, and the closer the subject is to the flash, the greater the f/stop.

Any doubling of the film speed will increase the f/stop output by one stop (and remember, the "full-stop" increments are f/2.8, f/4.0, f/5.6, f/8, f/11, f/16). If you halve the distance between the flash and the subject, you increase the f/stop by 1 full stop. If you double the distance between the subject and the flash, you decrease the f/stop by one full stop.

Now, for a particular flash power and ISO, (and distance too, just because of flash spread) this relationship between the distance and the f/stop is called the "guide number" for your flash. The GN is equal to the f-stop times the distance between flash and subject.

GN = f-stop x distance

Your flash should come with a guide number table, that tells you the GN for different film speeds and flash powers. For instance, your flash might have a GN of, say, 64 at ISO 100 for it's "full-power" setting, so in this example where I wanted f/16, I would divide 64/16 and get 4ft. So my flash has to be four feet from the subject in order to illuminate it at f/16.

Some of you might now be saying, "Hold up a sec Radlinski. Didn't you say the exposure under the rock was f/2.8 already? And now you're adding this flash to it, at f/16! So the correct exposure under that rock will be f/16 plus f/2.8, fool!" That's true. The proper exposure is the ambient light exposure plus the flash exposure. But let's look at what f/16 + f/2.8 means.

F-stops are logarithmic functions. That means when going from one f-stop to the next, we're dealing in powers (exponents) or two, not, say, multiples of two. If at f/2.8, it requires a certain amount of light to expose a scene properly, then at f/4.0, it requires twice as much light as at f/2.8. And at f/5.6, it requires twice as much light as at f/4.0, or four times as much light as f/2.8. So you can see where this is going.

If the exposure under the rock had been, say, f/11, and I added a flash set to f/16 there, that would change my exposure dramatically. f/11 is half as much light as f/16, so f/16 + f/11 = f/16.5. In that case, I would have set my flash to f/11 instead of f/16. f/11 (ambient) + f/11 (flash) = f/16 (total). But f/2.8 is five stops less than f/16, which it makes it 1/32nd as much light. So f/2.8 (ambient) + f/16 (flash) = f/16.03, which is essentially f/16.

And finally, this is all just theory. In the real world, I'd have a vague idea of the power of my flash from experience after initially gaining familiarity using the guide numbers, I'd put my flash about where I think it should be for a given ISO, power level, and f-stop, and then I'd run up to the subject with a light meter, fire off a test shot, and adjust accordingly. If the meter reading said I was underexposed, I'd either increase the flash power or move the light closer. If it said I was overexposed, I'd decrease the flash power or move the light farther away.

And finally, finally, all we're talking about here are the general mechanics of exposure and quantities of light. What actually makes a pleasing image (or an image that communicates whatever it was we wanted to say to begin with) is quality of light. Just because you're using the correct amount of light in a scene, doesn't mean it's being used in a way to produce the desired artistic effect you want. So while the exposure is important, it's also important to consider the direction of light, and the quality (i.e., softness or hardness) of the light. And modifiers that you add to change the quality of the light, like umbrellas or grid spots, are going to effect the guide number of your flash anyway, which is why I always meter and/or check the back of my camera.

Hope that helps

Matt

matt is absolutely right in his posts, but as a side note there are times when cameras have meters which are a bit off for one reason or another (usually QC)... You can shoot a test chart in controlled settings just to make sure your meter is accurate.

matt... you are a light nerd. And I'm lovin' it.

I can almost here you push your glasses up on your nose.

Let's all please keep asking Matt questions that produce these wonderfully detailed and lucid explanations. I plan on compiling everything here and publishing them if Matt doesn't get around to doing it first.

LOL.

+1! Matt is the man!

I have a question, Matt. Doesn't the matrix (in my Nikon) or evaluative meter try to compensate automatically for things like the sky? I read that it knows for example, which way up the camera is, and that if there's a very bright patch at the top it will de-sensitise that area. The zone system, in other words. They can also do things with focus distance and stuff to try to determine the actual subject of most importance.

I'm not saying it does a good job of it, but that's what it tries to do. Is the above advice therefore more relevant to center-weighted, or even full frame metering? I mean, the matrix meter does wierd and wonderful (and allegedly sophisticated) things. If we're seizing control like this, is it better to use metering that's predictable/understandable?

Also, your use of off-camera flash seems great. How do you do it? pocket wizard on camera, one stuck to the flash on a portable stand? Are you using it in this manual way the whole time, or do you for example, switch to TTL when you put it on the camera?

Thanks for your excellent tips!! Especially the f16.03 point. That was something I've wondered about for a while!

I just have to say this - Matt you are awesome! I'm saving this post so I can read it thoroughly later