Exposing to the right is one of the best ways to set yourself up for success in your color grade. But there is one critical detail that many of us overlook. Let’s take a look inside DaVinci Resolve and see if we can get a better understanding of exposing to the right.
Make some noise
To see the benefits of exposing to the right, let’s look at a series of left, right, and middle-exposed images. We have five images in our timeline, and I have exposed each image differently and then normalized it in Resolve.
The middle image, image three, is exposed to the middle of the exposure range. When I say “middle,” I mean that the image has a normal, healthy exposure that is neither underexposed nor overexposed. What I want to focus on today is image texture and noise, so let’s zoom in on this image to see the texture change as we go through different levels of exposure.
Comparing shot number three to shot number four.
As you can see, our normal “middle” exposure has some noise, but it’s not an excessive amount. Using the slider above, compare the middle exposure to shot number four, which I exposed one stop to the right. If you look at the lampshade to the right of the subject’s face, you can see there is not as much noise in shot number four compared to shot number three. That’s because when we expose to the right, we get less noise than if we have a middle exposure.
Pull it back
Let’s look at what we’re doing in Resolve to normalize our exposure. I exposed shot number four to the right and then used the node tree to pull the exposure back by one photometric stop to compensate. When we’re talking about exposing to the right, we’re not saying that brighter images are better than darker images. Rather, we are saying that exposing images to the right gives us some extra benefits when we get to the color grade.
If we look at shot number four, we can see there is a net change in texture and noise when we expose to the right and then pull it back to the middle. Let’s see what happens when we expose two stops to the right and then normalize it in Resolve.
Comparing shot number four and shot number five.
Comparing shot number three to shot number five.
In this case, there is even less noise the more we expose to the right. This is easier to see if you use the sliders above to compare all three shots. As you can see, shot number five is a cleaner image in terms of noise than shot number four, and it is much cleaner than shot number three, the middle-exposed image. This technique gives us more flexibility in our color grade. A colorist can do more with an image exposed two stops to the right than they can with a middle-exposed image.
Exposing to the left
Let’s look at a dramatic example of what happens when we expose an image to the left. Shot number two has been exposed one stop to the left, and I have then added a stop in Resolve to normalize the exposure.
There’s been a change in the texture of our image. We are now getting more noise in shot number two than in our middle exposure. Next, let’s go to shot number one, which was exposed two stops to the left and then normalized in Resolve.
Shot number one has quite a bit of noise, doesn’t it? To get the full dramatic comparison, let’s compare shot number one to shot number five.
Comparing shot number one to shot number five.
Which image would you rather use as a baseline for your color grade? I think most of us would choose shot number five over shot number one.
The noise floor
There’s another big difference between shot number one and shot number five. Notice what happens when we try to dig out some detail from the subject’s hair.
We can pull information out of the subject’s hair in shot number five, right? It’s simply a matter of what to dig out while maintaining the overall look of the image. But what happens when we try to do the same thing with shot number one?
There’s nothing in there but noise! We are at the noise floor of our camera’s sensor. The noise floor is where the merit of exposing to the right comes in. When we expose to the right, we are getting off of the noise floor of our sensor.
Imagine for a moment that we had a sixth shot in our timeline, and that shot was just a lens cap over the lens. What would we see? You might think we would see pure black, but it’s not that simple. What we would actually see is noise. Specifically, we would see the noise floor, which is the standing level of noise in the sensor.
The noise floor is what we see when our sensor is receiving no image or no light. That noise floor is what we’re trying to move away from when we expose to the right. If we expose to the left, too little light hits our sensor, and the result is noise.
Exposing to the right on set
Now that we know why we want to expose to the right, let’s talk about how we might do this when we’re shooting on set. Imagine that we are on set photographing shot number three. We would first want to configure our lights and camera, and then we would expose the image in such a way that we were happy with how it looked. That would be an example of exposing the image to a normal, healthy middle.
Now imagine that we want to expose to the right because we know that doing so can benefit our color grade. How do we do it? Well, there are multiple things that we could do. One way would be to change the amount of light hitting our sensor. We could do that by increasing our key light, fill light, or the overall lighting level in the scene. Another option is to open up the stop on our lens. That would be two ways of admitting more light onto our sensor, which is how we expose to the right.
The ISO mistake
But there is a third strategy some people try to use to expose to the right, and that is to raise the ISO. It sounds sensible on paper, doesn’t it? If you raise your ISO, you get a brighter image. We can prove that by copying our “+1” node from shot number two and pasting it onto shot number three. This is essentially what you would see if you were to change the rating of your camera from 800 ISO to 1600 ISO.
But changing the ISO is not the same as exposing to the right. That’s because when you increase ISO, you are not also increasing the amount of light hitting your sensor. Increasing ISO only brightens the image by what’s called “gaining” the signal. If we look at the Primaries, we can see that I am adding gain in a linear tone curve, doubling the amount of light, which increases the exposure by one stop. But I’m adding gain to a signal that has already been captured.
So what’s going to happen if you use this ISO strategy on set? There are a few possibilities, and none of them are great. The first possibility is that you will no longer be able to evaluate and connect with your image because it is now brighter than you ultimately intend it to be. How are you supposed to evaluate the fine nuances of your image if it looks too bright? Trying to expose to the right using ISO causes you to lose the relationship with your scene because you are looking at a reproduction that does not meaningfully represent what the final image will look like.
And honestly, that’s the best-case scenario. The worst-case scenario is that you could look at this brighter image and think, “Hey, we don’t need all of this light. Let’s stop down or take some light away.” That would be a very understandable thing to do, but it’s arguably an even worse scenario because you are now exposing to the left. Doing this causes less light to hit the camera sensor than if you had just stayed at 800 ISO.
Let there be light
The only correct way to expose to the right is to increase the amount of light hitting your sensor. That’s it. There’s no other way to do it. Luckily, there are many ways to increase the light. You can open up the iris in your lens or increase your shutter angle. You could also brighten your scene by adjusting the on-set lighting. Either way, you have to increase the amount of light hitting your sensor.
ISO cannot expose to the left or to the right, but it can add bias to your decisions. If you want to use ISO to aid in exposing to the right then you actually need to lower your ISO. If you lower your ISO and then compensate by increasing the amount of light hitting your sensor, you are using ISO to bias yourself into adding more light to your sensor as opposed to less.
