What bitrate measures
Bitrate is the amount of data used to represent one second of video (or audio). It is usually written in kilobits per second (kbps) or megabits per second (Mbps). A 5 Mbps video uses, on average, five million bits for every second of playback.
It helps to see what bitrate is not. It is not resolution, which counts pixels in each frame, and it is not frame rate, which counts frames per second. Those describe the picture. Bitrate describes the size of the compressed stream the codec produces for that picture. A 1080p video can be encoded at a low bitrate or a high one, and the difference shows up as how much detail survives compression.
How bitrate relates to file size
Because bitrate is bits per second, file size follows directly from it:
file size in bits = bitrate in bits per second x duration in seconds
Divide by 8 to get bytes. A worked example: a 10-minute video at 8 Mbps.
- 10 minutes is 600 seconds.
- 8 Mbps is 8,000,000 bits per second.
- 8,000,000 x 600 = 4,800,000,000 bits.
- Divide by 8: 600,000,000 bytes, or about 600 MB.
Audio adds a little more, and containers add small overhead, so real files are slightly larger. Using decimal units here (1 MB = 1,000,000 bytes); some software reports binary units, so the number shown may differ a little.
The same arithmetic works for delivery. Streaming at 5 Mbps for one hour moves 5,000,000 x 3,600 = 18,000,000,000 bits, which is 2.25 GB. That is why bitrate is central to both storage planning and the cost of streaming at scale.
What bitrate does to quality
When a codec compresses video, it must decide where to spend its bits. With a generous budget it can keep fine texture, gradients and motion. With a tight budget it has to discard detail, and the typical symptoms appear:
- blocky areas in flat regions or dark scenes
- smeared or blurry textures such as grass, hair or water
- banding in skies and other smooth gradients
- ringing or mosquito-like noise around sharp edges
These artefacts come from the steps that make lossy compression lossy. More bits generally means fewer artefacts, but the relationship is not a straight line.
Diminishing returns
Quality gains from extra bitrate shrink as the bitrate rises. Going from a very low bitrate to a moderate one can transform the picture. Going from a high bitrate to a much higher one often produces a difference that few viewers can see, while the file size grows in direct proportion. Past a certain point you are mostly storing detail the eye cannot distinguish.
Why the same bitrate looks different
There is no universal “good” bitrate, because the bits needed depend on several things.
- Content. A static talking-head video compresses easily. Fast sports, confetti, rain or heavy film grain demand many more bits for the same apparent quality.
- Resolution and frame rate. More pixels and more frames per second generally need more bits, though not in a simple proportion. Higher video resolution and higher frame rates both push the requirement up.
- Codec. Newer codecs such as HEVC and AV1 can often match the quality of H.264 at a lower bitrate. A figure that is generous for one codec may be poor for another.
- Encoder and settings. Slower, more thorough encoder presets usually get better quality from the same number of bits.
- Dynamic range and colour. High dynamic range and higher bit depth can require more data to avoid banding.
For these reasons, treat any recommended bitrate table as a starting point, and compare results by eye on your own footage.
Video bitrate versus audio bitrate
The figure people discuss is normally the video bitrate. Audio is encoded separately and typically uses a far smaller share, often a small fraction of what video does. Total bitrate is the sum of both streams, plus a little container overhead.
Bitrate and streaming
For streaming, bitrate has a second job: it has to fit through the viewer’s connection. If a stream needs more bandwidth than is available, the player runs out of data and pauses, which is one of the buffering causes. Services therefore prepare several versions of a video at different bitrates, and the player switches between them as conditions change, a technique called adaptive bitrate streaming. The set of versions a service prepares is often called one of its encoding ladders.
Peak versus average
Bitrates are often quoted as averages, but a stream’s actual rate moves up and down from moment to moment. Complex scenes need more bits and simple ones fewer. Whether the encoder holds the rate steady or lets it fluctuate is controlled by the rate control mode, which decides how far a variable bitrate may swing.
Practical guidance
- When exporting, think about the destination first. A file for upload to a platform that will re-encode it can be high quality, while a file for direct email or messaging needs to stay small.
- Avoid choosing the bitrate in isolation. Pick resolution and frame rate that suit the content, then set a bitrate that gives that picture enough room.
- Test short sections of your own footage at two or three bitrates and compare them on the screens your audience will use.
- Remember that re-encoding an already compressed video loses further quality, so keep a high-quality master when you can.