larryepage Loc: North Texas area

Today's people are not us. They are willing to learn new things, but they are not willing to get no usable results while learning. There a few exceptions, and I know a couple of those as well. But if it is too frustrating, most will just go back to their cell phone cameras, which are pretty formidable. The good news is that there are a lot of folks (not all, but many) that can do a really good job with those cell phone cameras. They have already learned about composition and framing and focus and holding the camera steady. So they can learn, when ready, to manage exposure. It's really the only initial missing link for them. Then they can learn to manage other technical aspects.

Let me ask this...how many here were taught to swim by being thrown into the deep end of the pool? That was still a technique being used when we were young. Three questions:

1. If so, did you learn how to swim? To swim well?
2. Did you enjoy learning? Do you still enjoy swimming today?
3. Do you really honestly think that is the best way to teach a person to swim?

OK...that's really five questions. But you still get the idea.

BebuLamar

Understanding Exposure by Bryan Peterson.

SuperflyTNT Loc: Manassas VA

I find it very ironic that your profile pic is a light meter.

Bipod

The "correct" exposure is the one that matches the photographer's visualization of the scene.

In average contrast scenes, there is usually "exposure latitude" -- he can be off a
stop or two and fix it later in processing--assuming nothing else goes wrong.

Exposure becomes crucial in constrasty scenes--when the number of stops separating the
darkest shadow and brightest highlight approaches or exceeds the dynamic range of your
camera's sensor. Detail that was lost cannot be restored in processing--any information
that the sensor didn't capture is gone forever.

Unfortunately, photography is not WYSIWYG ("what you see is what you get").
The human eye can only see 10 stops of contrast at one time, and does a poor job of separating
dark tones. For this reason, blown highlights (no detail visible) are always unrecoverable,
but shadows that appear pure black are often recoverable.

Unfortunately, it's not easy to determine the dynamic range of a scene unless you own
an accurate hand-held spot meter, but you can get a rough idea from the histogram in
live view mode. Also, camera manufactuers have been known to exaggerate the dynamic
range of the sensors used in their cameras (and that's putting it mildly). If you know which
image sensor is used in your camera, you can download the datasheet for that part number
from the chip manufacture--those are usually accurate.

Any camera setting that could cause the image file to depart from what you wanted is
important--but the most important are mistakes that cannot be corrected in processing.
However, processing is something of a "black box"--you click on a digital filter and
it does it does whatever it does to the data in your image file (you're not privy to the
details). So it's better to treat processing as a safety net rather than as an elevator.

If a digital filter is fully reversible, then you know it's not losing information.
For example, global color correction: you can go back and forth as many times
as you like without degrading the image. But some digital filters (e.g., "sharpen"
and "blur") are lose information: sharpen ==> blur ==> sharpen doesn't
produce the original image, it produces a very degraded image. Unfortunately,
most software packages don't bother to tell you which digital filters are
"information lossy".

For example, "sharpen" trades gradation for accutance and can create lines where
there weren't any in the actual scene (an example of a digital "artifact"). So focus
mistakes cannot really be "fixed". Worse, you may not notice the loss of gradation
in the small image on your monitor, so you'll be in for an unpleasant surprise when
you print the final image.

LCD/LED monitors have even less at-a-time contrast than the human eye. And prints
have least of all. The appearance of simplicity --- "just process the image until it looks
good on your montior"--is deceiving. Photography has always involved planning for
the final image (whatever medium it will be) and one needs to think ahead and take
measurements--not trust one's eyes.

Anyone who's been on a Hollywood film set will have noticed the frequent use of
incident light meters to assist in determining exposure and tape measures to
determine focus distance. Mistakes are really expensive, and cameramen and even
DPs have been fired from a film for making them.

BebuLamar

Because really that dude didn't teach about the light meter. Only the triangle which has nothing to do with measuring light. Read the book and you will find he talked very little about how to measure light. Very little about how to determine exposure.

Bipod

SuperflyTNT Loc: Manassas VA

Have you read the book? As I remember he talks a fair amount about where he’d meter different scenes. As a matter of fact, one of the complaints I’ve heard about his book is that they wish there was less about metering and more about reading histograms for exposure. And it still doesn’t answer why you think the exposure triangle is “fake”.

Bipod

If I may interject: the process of determining exposure is a dialog between the photographer
and whatever measurements are available (light meter, time of day, wether, etc.).

The primaryquestions is "what exposure challenges does this scene present"?
Typical answes would be "backlighting" or "contrasty", etc.

It all starts with understanding metering. Light meters are not calibrated like scales or thermometers.
You might want to take a look at this:
http://dpanswers.com/content/tech_kfactor.php

The issues are well-developed in the linked article, but the conclusions are wrong.
IMHO, light meters measure a physical scalar quanity. Therefore, they should
all agree, regardless of brand--just as thermometers do. Ansel Adams was correct,
Kodak was wrong.

If camera manufacturers want ot introduce their own metering constants--fine, they can do
so as part of the auto-exposure algorithm. But all voltmeters, all thermonmeters and all light
metsrs should agree--regardless of brand. Currently, light meters do not, because of the
'k factor".

The units of measure for luminance and illuminence are a giant mess and need to
be cleaned up. Until then, the best thing we can do is to stick to SI units: candela
pers sq. meter for luminence and lux (= lumens per sq. meter) for illuminance..

"Exposure value" is much handier for photography, but unfortunately EV does not
seem to be defined by any internaional standard.

Sorry this is so complicated. If corporations and governments were made up of
scientists, it wouldn't be.

BigDaddy Loc: Pittsburgh, PA

Not sure why he says it's fake, but perhaps the triangle today is really 1/2 a rectangle? If true, and it probably is, then at least 1/3 of the triangle is "fake", at least in digital photography.

SuperflyTNT Loc: Manassas VA

1/2 of a rectangle? How is digital any different than film? Exposure is exposure.

BebuLamar

I have read the book. The triangle is simply a stupid thing. He spent 3/4 of the book talking about the triangle yet only a couple of page about metering.

AndyH Loc: Massachusetts and New Hampshire

The electronic sensor performance, measured by ISO, is different in many ways from film speed. Reciprocity, color response, ISO invariance, etc. We just use ISO as a convenient reference - it is NOT a measurement of how the sensor responds to or receives light, so much as how the camera electronics manipulate that given amount of light. There is no "triangle" the so-called "third leg" is just a little computer that responds to light. That it responds differently can be seen in ISO invariance, exposure to the right, noise development (very different from grain development) and other problems that are basically electronic. Stick to getting the right amount of light to the sensors simplifies thinking about exposure.

Andy

SuperflyTNT Loc: Manassas VA

So let me ask you a question, do you either set the ISO or use auto ISO when shooting digital? Maybe ISO is just a “convenient reference” but even so, ISO invariance is not 100% yet. No camera is ISO invariant throughout its range. And I would conjecture that pushing in PP is really just an extension of that side of the triangle. Like pushing film speed when developing. On the flip side, if you’re an ETTR proponent you’re certainly using ISO as one of your settings. And it’s crucial if you’re an SOOC proponent. I will say that if you want to get taken back to the days of film, by shooting everything at low ISO and pushing it in processing you will get that anticipation of not knowing what you have until you get back to the darkroom, (or Lightroom these days), no more chimping.

AndyH Loc: Massachusetts and New Hampshire

I set the ISO for the lowest level I anticipate might work in the circumstances and light where I'm shooting. That's usually 100 for outdoors, as much as 6400 indoors on my D7100. If I can't get a combination of shutter speed and aperture that makes me happy, I will first manually dial up the ISO. In metering, I am spot metering the high values and working down from there - I can more easily bring the shadow values up in PP than I can find some details in blown highlights where there is no data. If I get too frustrated or am switching too quickly back and forth between high and low value areas, I will sometimes float the ISO, with pre-set limits. I almost always chimp, looking for blinkies. If they're too big, or on areas where I want texture, I turn the exposure down until they're only covering specular highlights.

This is basically exposing to the right, and counting on your post processing software to spread out or compress the values.

Anyone else can do what they like - I only care about what works for me. And that's a combination of spot metering on the highlights, keeping ISO as low as possible, and working with RAW files in post processing. If the "exposure triangle" works as a metaphor for you, I'm happy, but it has very little to do with exposure, it's more a metaphor on how to view setting the controls on your camera.

Plenty of people seem to love this metaphor, and I'm all for it if it helps you work. But I personally see it as part of dumbing down photography to the next level. If it helps you take better images, it's still a good thing! :D

Andy

BebuLamar

You only outlined one technique in a few paragraphs but it's more than the entire book "Understanding Exposure" has about exposure in itself. The whole book talking about the triangle which involves choosing the aperture, shutter speed and ISO only after the exposure has been determined.