How to Record High-Quality Video for Colour Grading

If you’re a filmmaker and you’d like to know more about editing, particularly colour, then you’ll our free course, How to Colour Grade Video. In this lesson, you’ll learn how to get the most from your colour grading by shooting your footage in the best way.


Getting the most out of your colour grade starts with good footage, and in this lesson, you’ll learn what that looks like. First, you should know that not everything needs to be colour graded. Some projects don’t have the or budget to make grading a separate step in the process, and in those cases, what you want to do is get it to look the way you want, in camera.

If it looks perfect in camera, all you really need to do is edit it and encode it. You won’t have the latitude to do wild colour manipulation with that kind of footage, but if you aren’t intending to grade it anyway, it won’t matter.

How to Record High-Quality Video for Colour Grading

For footage that you do intend to grade, you’ll need to take a different approach. Post-production video is an RGB world—the colour of each pixel is made from the additive process of mixing red, green, and blue. Everything in colour grading and colour correction is essentially re-.

To warm up your image, you’d be increasing the exposure on the red channel a lot, the green channel a little, and decreasing the exposure in the blue channel.To warm up your image, you’d be increasing the exposure on the red channel a lot, the green channel a little, and decreasing the exposure in the blue channel.To warm up your image, you’d be increasing the exposure on the red channel a lot, the green channel a little, and decreasing the exposure in the blue channel.
To warm up your image, you’d increase the exposure on the red channel a lot and the green channel a little, and decrease the exposure in the blue channel / David Bode

If you want to warm up your image, for example, you’d be increasing the exposure on the red channel a lot and the green channel a little, and decreasing the exposure in the blue channel.

If you want to bring up the shadow detail, you’d be re-exposing the lower values in each one of the RGB channels.If you want to bring up the shadow detail, you’d be re-exposing the lower values in each one of the RGB channels.If you want to bring up the shadow detail, you’d be re-exposing the lower values in each one of the RGB channels.
If you want to bring up the shadow detail, you’d be re-exposing the lower values in each one of the RGB channels. / David Bode

If you want to bring up the shadow detail, you’d be re-exposing the lower values in each one of the RGB channels.

To give your shadows a blue tint would mean increasing the exposures of the lower values in the blue channel while reducing the values in the red and green channels.To give your shadows a blue tint would mean increasing the exposures of the lower values in the blue channel while reducing the values in the red and green channels.To give your shadows a blue tint would mean increasing the exposures of the lower values in the blue channel while reducing the values in the red and green channels.
To give your shadows a blue tint would mean increasing the exposures of the lower values in the blue channel while reducing the values in the red and green channels. David Bode

You can do this equally or individually. For example, you might want to give your shadows a blue tint, and this would mean increasing the exposures of those lower values in the blue channel while reducing the values in the red and green channels.

It’s important to know this because the more detail you have in your image, the more you can do with it. If you don’t have detail in the shadows when you try to re-expose them, it’ll look terrible. Footage with fewer details means you have less room to push exposure values around before the image starts to fall apart.

What Does ‘Detailed’ Footage Look Like?

An image with a lot of contrastAn image with a lot of contrastAn image with a lot of contrast
An image with a lot of contrast / David Bode

In the still from a clip, above, you can see there’s a lot of contrast. Most cameras will record this image by giving the majority of the compression bandwidth to the brighter areas and very little bandwidth to the darker areas, so if you try to push up the shadows, it becomes a real mess.

Pushing up shadows can lose a lot of detail and look badPushing up shadows can lose a lot of detail and look badPushing up shadows can lose a lot of detail and look bad
Pushing up shadows can lose a lot of detail and look bad / David Bode

Very little bandwidth, or data, has been allocated to that area of the image.

Log footage retains detail where neededLog footage retains detail where neededLog footage retains detail where needed
Log footage retains detail where needed / David Bode

This is a logarithmic gamma, usually called Log. Instead of a lot of contrast, we get an image that has brightness values mapped so the darkest parts of the scene appear lighter and the brightest parts of the scene appear darker, and this puts more of the detail in the zone where the bandwidth is.

The 'flat' log footage (right) retains more details than the image with more contrast (left)The 'flat' log footage (right) retains more details than the image with more contrast (left)The 'flat' log footage (right) retains more details than the image with more contrast (left)
The ‘flat’ log footage (right) retains more details than the image with more contrast (left) / David Bode

The result is a picture that looks flat and lifeless, but it actually has more usable information compared to the image with a lot of contrast. Log also works really well for cameras that record in a lossless, compressed format, or Raw.

With Raw, nothing gets thrown away, so you get everything that the sensor captured. Given the options, Raw would seem like the most logical choice, but there are some drawbacks. Raw files have to be converted in order to be used in your editor, and that’s extra time that you might not have. The file size can be massive, too, and that can be problematic in terms of storage and processing.

Choosing the Right Profile

Choosing the right profile to shoot in will help get the best footageChoosing the right profile to shoot in will help get the best footageChoosing the right profile to shoot in will help get the best footage
Choosing the right profile to shoot in will help get the best footage / David Bode

The main thing you can do to get the most out of your camera is to set it up so that it’s capturing the most detail. If you have a Log picture profile or picture style, you should use that. You might even have a few log profiles to choose from, and it’ll take some experimentation to figure out which one will work best for what you’re shooting.

If you haven’t worked with Log footage before, you’ll need to make sure you have a method for setting exposure accurately. Because Log looks so flat, you may want to adjust your exposure up or down, and that can be dangerous because if you expose too far down, it’ll be quite problematic trying to recover all of that information. If you expose too far up, you may be clipping things that you didn’t mean to, so that’s something to watch out for.

Something else to consider is how you’ll store your video on your camera or what kind of codec you’ll use to record the video. You want the most amount of detail, but within reason. If you have the option to shoot Raw or ProRes, then Raw might be a lot of extra work—codecs like ProRes and DNxHD with Log picture profiles can work just as well as Raw, with far less hassle. That’s because they can be edited without transcoding or converting, and their performance is very fast.

Alternative Shooting Settings: No Profile/Codec Options

If you don’t have the option to shoot in one of those higher-end codecs like Raw, ProRes, DNxHD, or CineForm, you should record to the highest bitrate possible because the codec will lose a lot of the visual information.

But maybe you have a camera system that doesn’t have those high-end codecs or the option to use a Log picture profile. In that case, record with your profile as flat as possible, which will mean going into your camera settings and tweaking the picture to reduce the contrast and saturation and turn off sharpening. This will get the picture close to what a Log profile would look like because a decrease in the contrast and saturation will lift the blacks and lower the highlights quite a bit. The image will look dull and lifeless, but you’ll get far more detail in the zone where the bandwidth is, and this will make colour grading a lot easier.

Exposure

Exposure is another thing to consider when you’re shooting for a grade. As we touched on earlier, very little bandwidth is given to the darkest parts of the image, so if there’s shadow detail that you want to see, you’ll need to expose “to the right” if you can. This means if you have room to increase the exposure while you’re shooting—without clipping important details—you should. You can always darken an image in post-production in a very clean and subtle way, but if you’re trying to increase the exposure in the shadows, it usually looks worse.

If you are going to try to increase the exposure to capture more of that shadow detail, you should do it very carefully and avoid clipping your highlights too badly or clipping important information.

Summary

Use a Log picture profile if you have it and know how to expose for it. If you don’t have a Log picture profile, use an option to flatten out your picture profile on your camera by reducing the contrast and saturation and turning off sharpening.

Use the highest bit rate codec you can. For darker scenes, expose to the right where appropriate—and watch you don’t clip those highlights.

More Colour Grading Resources

About the Authors

David Bode created the video course that includes this lesson. Dave is an expert on video and audio production, and he lives in the upstate NY area. He works as a camera operator, editor, inventor, motion graphics designer, recording engineer, and studio musician.

Marie Gardiner wrote the version of this lesson, and it was edited and published by Jackson Couse. Jackson is a and the editor of the Photo & Video section of Envato Tuts+.