Technical Considerations When Buying a Video Camera (from Resolution to Bit-rate)

This article is a supplement to ‘How to Choose The Right Camera’ and covers the key technical terms involved.

Resolution

Resolution refers to the number of pixels in an image. Just like in photography, resolution determines the level of detail and how well an image will hold up when enlarged. Unlike photography, however, the resolutions we work with are much smaller to account for the vast number of frames per second.

The standard for web video is HD (high definition). In the UK, you’ll see ‘1080p25’. That’s 1920 horizontal pixels by 1080 vertical pixels and 25 frames per second

At present, the majority of online content is played back in HD. Over the next few years, the standard will inevitably become 4K, which is 4096x2160 pixels (roughly 4 times the resolution of HD). To confuse things a little, there is also a ‘consumer’ version of 4K, called UHD, with slightly fewer pixels (3840x2160).

Camera equipment always stays ahead of the trend when it comes to resolutions because of the extra time it takes the slower moving consumer market to upgrade their displays. Although it’s possible to record in 6K or 8K, there are very few viewers at present who would be able to appreciate it. As displays become more affordable however, the demand for 4K, 6K, and 8K will gradually increase.

Moving from HD up to 4K requires not only more expensive camera equipment and memory cards, but significantly more powerful computers to edit with.

As well as increased costs, there are also new technical challenges that arise with 4K. Keeping a sharp focus on a 4K image is particularly challenging, so you’ll need to pay extra special attention while filming.

Just like standard definition (SD) video, high definition (HD) will eventually become obsolete. It will be a while before every house has a 4K TV or monitor, though. If you don’t know what decision to make, consider this:

• If your project might continue to generate income for years to come, you should consider this an investment.
• If your budget could be allocated better elsewhere, then you don’t have to worry about this. At the moment, HD is still acceptable for most applications, anyway.

Sensor Size

A digital sensor is the electronic light-capturing device at the front of the camera that converts light into a digital signal. When choosing a camera, you’ll need to understand the various sensor sizes available, as each one will affect your image differently. The sensor is in the camera body itself and is one of the main price differentiators.

Although bigger sensors tend to be more expensive, it doesn’t mean that they’re always better. There are a few factors to consider that will help you decide which sensor is best for you.

Crop Factor

Below is a comparison of different sensors and their ‘crop factors’, which determine the resulting focal length when a particular sensor is paired with a full-frame lens.

When pairing a full-frame sensor with a full-frame lens, the focal length stated on the lens (e.g. 50mm) will be true to size. When you work with cropped sensors, like the popular Canon APS-C on full-frame lenses, you introduce ‘crop factor’. This impacts the equivalent focal length of your images. For example, the 50mm lens on an APS-C sensor becomes the equivalent of an 81mm (50mm lens multiplied by the 1.62 crop factor).

When working with cropped sensors, it’s important to remember that your focal lengths will not be the same! This can become troublesome when working with high crop factors, like the popular micro four thirds cameras (with a 2x crop factor). On one of these sensors, a 24mm full-frame lens becomes 48mm, which is very impractical for a wide shot. You’d need a 12mm lens to get the equivalent shot.

There are purpose-made smaller lenses for sensors like the micro four thirds that will give you the ‘correct’ focal length. However, if you already own full-frame lenses, then micro four thirds is a whole new investment. You won’t be able to use smaller lenses on larger sensors, e.g. micro four thirds glass on a full-frame body.

Depth Of Field

​The size of the sensor will also affect your depth of field (the distance in front of and behind your point of focus that is perceptibly sharp). Larger sensors are known for having a desirably shallow depth of field, which creates a filmic look. This is one of the main reasons that the full-frame Canon 5D series revolutionised the industry. Despite its low price, it has been used a lot in big-budget film and television shows.

Note: Technically, smaller sensors have a shallower depth of field. However, when working with full-frame lenses, you need to be closer to the subject on a larger sensor to achieve the same focal length and framing (remember crop factor). Hold your hand in front of your face and move it backwards and forwards. The closer it is to your eyes, the softer the background becomes.

Low Light Performance

When choosing a camera, you need to consider what lighting conditions you’ll be working with the most. If you’re shooting in daylight or have lots of lights available, then this won’t be such a concern. If you’re planning on filming at magic hour, at night, or in dark interiors, then you’ll need to explore specific sensors for this purpose.
​
Most DSLRs become too grainy beyond ISO 1600. However, if you’re presented with two different size sensors with the same number of megapixels, the larger one is likely to work better in low light. A recent breakthrough in this area was the Sony A7Sii, a compact low-light camera that has a usable ISO range up to around 80,000.

Bit Rates & Image Quality

Up until now, we’ve looked more at video dimensions than video quality. There are multiple factors that determine the image quality of a video, and bit rate is one of the most significant. Bit rate is measured in mega bits per seconds (Mbps). If all other factors remain constant, you get this formula:

higher bit rate = higher quality

To put bit rate into perspective, the recommended export settings (as per YouTube’s guidelines) are 2.5 Mbps for for a standard definition video, 8 Mbps for HD, and 35-45 Mbps for 4K.

When choosing a camera, it’s important to take into account what the bit rate is relative to the resolution. If you’re planning on colour grading your image, then you’ll need to make sure the bit rate your camera is recording at is suitable. Even though you may export your HD video to just 8 Mbps for YouTube, you’ll want to record at least 25-35 Mbps.

As you increase the bit rate, you increase the file size. It’s generally not an issue with HD, but when you move up to 4K, you’ll want to budget for adequate storage space. This is especially important when it comes to projects like documentaries, where you’re likely to collect more footage. An HD image at 35 Mbps may take up around 15 GB an hour, whereas a 4K image at 100 Mbps is closer to 45 GB an hour.

Here are a few recommended bit rates for recording. They will also leave room for minimal colour correction:

SD 480p – 10 Mbps – 4.5 Gb/hour
HD 1080p – 35 Mbps – 15.8 Gb/hour
4K 4096p – 100 Mbps – 45 Gb/hour ​

Think of your image as a ball of pizza dough. The data rate would be the density, and the resolution would be the quantity. If you stretch out poor quality dough then it’s going to fall apart, so you would be better off making a smaller, higher quality pizza

Bit Depth

Bit depth tells us the number of bits used to represent the colour of a single pixel. The terms you will see here the most are 8 bit, 10 bit, and 12 bit. To put it simply, a higher bit depth means more colour information. An 8-bit image is the standard for most consumer (and some semi-professional) cameras. It looks acceptable when played back, and it’s something most of us are used to watching both online and on TV.

When shooting video, you need to be aware that there isn’t nearly as much colour information available in an 8-bit image compared to 10 or 12 bit. This is particularly important if you want to colour grade.

Where 8 bit is 256 tones per colour channel (over 16 million possible colours), 10 bit is 1024 (over a billion possible colours). At 12 bit, you have 4096 tones (over 68 billion possible colours).

Now, 16 million colours sound like a lot, but you can distinctly see the difference between 8 bit and 10 bit. If you’re working with an 8-bit image, I recommend using more of the ‘ready-to-go’ picture profiles for a better overall result. When moving up to 10 bit, I will almost always shoot in the flat LOG profile because of the increased information and flexibility for grading.

10 bit and 12 bit colours can look amazing, but most entry-level cameras don’t support them. More often now, some cameras that record 8 bit internally can output a 10-bit image to an external recorder. This is a useful feature because it gives you the option to either buy or rent an external recorder only when you need the added depth. Higher bit depths use more storage, and you may not need or want to deal with the extra load all the time.

Whether you need 10 bit or not will depend on your colour grading and displaying plans. Although most consumer screens don’t yet support 10-bit images, compressing down to 8 bit will still look significantly better than an image that started life at 8 bit.

Bit Depth	Tones Per Colour Channel	Possible Colours	Use
8 bit	256 tones	16 million	Consumer grade. Limited flexibility but easy to manage
10 bit	1024 tones	1+ billion	Professional grade. Good balance of flexibility and size
12 bit	4096 tones	68+ billion	High end with extreme quality and files sizes

Chroma Subsampling

Chroma subsampling is another step in the compression process. It’s different to bit depth, and has less of a visible impact on your image, but it’s still important to consider, especially if you plan on colour grading. To save storage space, pixels are grouped together and are made to share colour information with one another. The result is very effective, but we can hardly tell the difference because the human eye is a lot more perceptible to changes in brightness than changes in colours. These are the most common options:

4:4:4
With no sub sampling applied to the image, this is the highest quality option. Every pixel has its own luminance and chroma value. As a result, you end up with very large file sizes. Only high end cameras can record or output 4:4:4.

4:2:2
This reduces the chroma information by 50% to save on storage space while offering a good amount of information for colour grading. You can find this option internally in only a few sub £10,000 cameras, but more commonly in this price bracket when outputting to an external recorder. 4:2:2 is also associated with being the standard for broadcast television.

4:2:0
This reduces the chroma information to about 25% of 4:4:4. It’s the most common of the three in consumer and semi-professional cameras. It’s efficient with storage space but doesn’t hold up so well for colour grading.

For most entry-level cameras, you’ll find 4:2:0 is your only option. If you’re taking your colour grading seriously or working with green screen, then the extra colour information from 4:2:2 or 4:4:4 becomes valuable. Many broadcasters also wont accept below 4:2:2.

Raw

Some high end cameras give you an option to skip compression and record in a ‘raw’ format. By taking data straight from the sensor you get a very high quality, uncompressed image that hasn’t been affected by the image processor. This data has no set colour temperature, ISO, or subsampling applied, giving you a lot more information to work with in post production.

The raw data that’s captured is in fact so raw that you can’t play it back with your usual media player. Cameras that have raw capabilities come with software to read or convert the raw files into something you can view. This process adds a lot of time to the editing workflow.

When it comes to raw, you get huge file sizes. Also, a lot of computing power is required to handle and play back the content. You’re unlikely to have a use for raw unless you’re working on a high budget production or have very specific technical requirements.