IAN: All right, folks.

Welcome back to week six, where we're going

to talk a little bit about exposure.

So to this date we've talked a lot about composition, lensing, storytelling,

without focusing too much on the technical details of how

to actually make an image, and match it to our intentions.

So we'll dive a little bit more in that today.

And I think it's important to define exposure.

So the idea of exposure is that we want to render

a scene in a specific way using our camera controls to interpret

the amount of light in a given scene.

So what does that really mean?

Well we're going to dive deep into this, and we're

going to look at each of the different camera controls.

How they affect the image, and how we can utilize

them to make different exposures.

So what's our goal with exposing an image?

It's really to capture that intentional image.

It's to make a decision about how we want an image to look

before we press the shutter button.

When you have a camera that's on auto mode and you press the shutter button

you're basically giving up all the decision making to a small machine.

And they're sophisticated, sure, but they're not intelligent.

And our goal is for us as intelligent operators

to make the decisions for the camera, or at least

override decisions that may be poor.

So I think what we really need to understand

is what exactly is it that we're trying to do?

And so in any given scene there's some amount of light.

And we can use absolute measurements to tell how much that is.

It could be 150 foot candles of light, but I don't know what that means.

I sort of conceptually understand that, but it doesn't help me take a camera

and make an image.

I have to do a bunch of math.

It gets very confusing.

There's easier ways.

So what we really need is a relative amount of light,

or at least the scale--

a relative scale that we can use to adjust our camera settings.

So for us in this class when we're talking about photography,

and then later cinematography, we're going to use the concept of f-stops.

And all an f-stop is is a doubling or halving unit.

So if you have a camera that has some sensitivity,

and you double the sensitivity of it, that's one stop.

If you have a camera that has some light coming into it,

and you cut the amount of light coming into it by one half, that's one stop.

They're going in opposite directions, but it's

an equivalent unit of one stop.

So just briefly to get a sense of how this

might look when we look at an image we have here an image from Mount Auburn

Cemetery that's at exposure, roughly.

And I made this exposure using a DSLR, and the light meter

that was in the camera told me that if I set the camera settings to this

it'll be at exposure.

And I think it did a pretty good job.

There's no real complaints there.

So in an effort to sort of investigate how much difference a stop might be

I opened up one stop.

So I allowed double the amount of light to strike the sensor.

And you can see that it gets much brighter.

If we go backwards we have some dark shadow detail in here, neutral gray,

some bright white, and everything feels naturalistic.

We start to brighten up that neutral gray begins

to push towards the lighter gray, this definitely begins to clip,

these trees feel a little bit brighter.

Not completely unnatural, but getting there.

So now if we allow four times as much light in, or two stops--

remember that you're doubling each time you open up a stop.

So you double, and then you double again.

Now this image really starts to fall apart.

And this is the classic overexposure that maybe you've accidentally done,

or you've been struggling with your camera controls.

And you end up here where you have clipping elements, there's no detail,

there's no actual shadow detail.

No dark tones in this at all.

It's all light grays.

And if we go even further it completely falls apart.

And we could keep opening this--

allowing more light to hit the camera and eventually we'd

end up with a solid white image.

So, OK, that's overexposure.

But when we look at this image again--

add exposure.

This is the same image from the beginning.

And we go the other way, I've now reduced

the amount of light entering into the camera by a half, or one stop.

And you can see that all the tonalities get depressed.

They're getting pushed down into the shadow areas.

There's more detail in that white snow, which was much brighter before,

but now it's sort of a shade of gray.

And if we keep going two stops so four times less light

is reaching in the sensor you now see like beautiful detail in here,

but everything else is turning into a muddy, crushed, shadowy darkness.

And if we keep going it becomes almost unreadable.

So all that is is to say that we have these tools that

allow us to increase and decrease the amount of light,

and we do so by measuring it in doubling or having units.

So how do we measure light?

Well, we have a brief video online that we posted earlier this week, on Friday,

on light meters, which if you've all watched, great.

And if not we'll talk a little bit about as a refresher.

But there's a few different ways.

You can use a handheld light meter like this to measure light,

but I think more often than not what we'll use

is the internal light meter of the camera.

And so this will measure the amount of light that's striking the subject,

and reflecting back through the lens onto some sort

of light sensitive material.

And it will say if you set your camera to these values you'll

get a decent exposure.

So they can be hand-held or internal to a camera,

but all light meters are calibrated to expose

for an idea called 18% reflective gray.

And they do so in different ways.

So briefly, before we look at a couple different ways

they do that, 18% reflective gray is often sort

of colloquially termed middle gray.

When exposed properly it forms the middle value between absolute light

and absolute black in your image.

It's right in the middle.

So it's a very handy reference tone for us.

And it's usually found on a gray card--

something like this, that reflects back 18% of the light that strikes it.

Cool.

So before we get here let us quickly fire this guy up.

So this is the output of our camera right now.

If we take a middle gray card and we place it

in front of the camera well illuminated--

we'll fill the frame with it.

We can see that in the middle of this there's this histogram,

and sort of all of the image data is centered right in the middle.

If we set it for this white background you

can see that it's sort of drifted back to the middle as well.

And that's sort of interesting because we would

expect this tone to reflect more light.

So there is a bit of a trick going on with reflective light meters

that we need to pay attention to.

Reflected light meters always assume that every single element

that they're metering is middle gray.

So there's a problem when you meter something that is not

middle gray like this white background.

And so if we actually are to open up this camera

and allow a little bit more light in--

what am I doing here?

There we go.

I has auto ISO on.

So there's the shadow.

Let's get this shadow out of here so we can--

so now all of a sudden we brighten the image up by allowing more light in,

and this tone is rendering normally or as we would wish it to.

So we can also do the same thing if we have a black object.

You can see that the camera using the reflective meter

set this value to add exposure, which is lining up

that small cursor with the caret in that scale.

You can see that this is not black.

It's middle gray.

It's actually rendering this incorrectly because what it's expecting to see

is this middle gray card.

So what we need to do in order to offset that

is actually underexpose what the camera is reading.

So now when we check this the camera is like you're over three stops

underexposed, but this starts to look correct.

So we talked a lot about this-- thanks, Ben.

[INAUDIBLE]-- in the light meter video.

And we'll come back to this, but that is to say that in any given scene

the exposure suggested by a camera is made to be 90% accurate.

Most scenes have a mix of light values, dark values, and values in between.

So if you sort of assume that that all melds down to about middle gray

you can suggest an exposure.

And the meter will suggest an exposure, and then

when you expose for that you'll be pretty accurate.

But so this scene here is pretty much all highlights.

And my camera said if you set it to these settings

you'll get a properly exposed image, but it's not.

It's too dark.

So if we open up one stop.

Well, we're getting closer.

Snow starting to look like snow.

And if we open up one more stop maybe we're a little bit too far.

It's getting a little bit bright over in this area over here,

but it's rendering as a very bright white.

And sort of side by side you can see it here.

The metered exposure versus compensating by increasing our exposure one stop.

So this is what I mean to say when I say that we need to be a little bit more

intelligent than the machines.

They're very sophisticated.

They can measure all kinds of things, but they're still sort of locked in

and reference at specific values.

So we need to understand what those are so

that we can compensate, and sort of use our intention

to override the camera when it's making, essentially, a dumb decision.

Yeah?

AUDIENCE: If middle gray is meant to be halfway

between pure black and pure white how come it's 18% and not 50?

IAN: So it has to do with the sensitivity of--

or the way that human eyes render light, and because it's logarithmic.

So it's a power function.

So 50% is not quite halfway on a linear scale

when it gets transformed, I think, but essentially it

is because if you have say--

this is a good example.

If you have one light bulb, and you turn it on,

and you add another light bulb it's almost

like you've doubled the amount of brightness,

and it's a very obvious difference, but if you have 100 light bulbs

and you turn one more light bulb on it's such a tiny incremental difference.

So we're dealing with light on this power function scale

where we're doubling and we're halving, and so the value that gets us

to that middle gray is actually 18% and not 50%

because it's not a linear scale.

Which there's plenty of math, and I certainly won't do it justice,

but we can dive down that rabbit hole another time.

So we looked a little bit about what exposure is doing

and how we might fool ourselves with the light meter,

but what are the actual settings that I was just changing on the camera?

I was pushing and pulling at some buttons and dials,

and I was clicking some things, and the image was changing, but what was it?

So the three main camera controls.

The first one is ISO, which is the sensitivity of a sensor to light,

or a film stock, or any kind of medium that we're using to capture light.

The second is shutter speed.