What is video compression: the term explained
Video compression is the process of making a video file smaller by discarding the data a viewer is least likely to miss, so it can be stored cheaply and sent over a network at a practical size. Almost every video you record, send or watch has been compressed.
The reason it exists is arithmetic. Raw, uncompressed 1080p video runs to hundreds of megabytes per second of footage. Nobody is texting that to a customer.
What does video compression actually remove?
Two kinds of redundancy get squeezed out.
- Spatial redundancy. Within a single frame, large areas are near enough identical. A flat grey bonnet does not need every pixel described individually.
- Temporal redundancy. Between frames, most of the picture has not changed. The encoder stores one complete frame (a keyframe) and then, for the frames after it, records only what moved.
- Perceptual detail. Human eyes are far more sensitive to brightness than to colour, so colour information is stored at lower precision than the luminance information. You will not notice unless you go looking.
Mozilla's MDN Web Docs puts the trade honestly: most video codecs are lossy, meaning the decoded video does not precisely match the source, and as a general rule the more compression you achieve, the more loss of detail and fidelity occurs. Lossless video codecs exist, but they are used for archiving and local playback, not for sending anything to anyone.
Try Venta Capture on your own process
Build one flow for your highest-volume case type. Free, no credit card.
Codec and container are not the same thing
This is the distinction people get wrong most often, and it causes real support tickets.
A video codec is the compression method: the set of rules for encoding the picture and decoding it again. H.264, also called AVC, is the common one. MDN notes that AVC is specified by the identical ITU H.264 and MPEG-4 Part 10 specifications, which is why the same codec turns up under two names.
A container is the file format that wraps the compressed video stream together with the audio stream, any subtitle tracks, and the metadata. MDN defines it as a file format that encapsulates one or more media streams along with metadata, so they can be stored and played back together. MP4, MOV and WebM are containers.
So MP4 is the box. H.264 is what is inside it. Two files can both be MP4 and still behave completely differently, because the container tells you nothing about how hard the picture was compressed.
How bitrate, resolution and quality relate
Three numbers get confused constantly, and only one of them is quality.
- Bitrate drives file size. It is the amount of data used per second of video. Double the bitrate, roughly double the file. It is also the single biggest lever on how good the compressed picture looks.
- Resolution is not quality on its own. A 4K file starved of bitrate looks worse than a well encoded 1080p file, because there are more pixels to describe and no more data to describe them with. Resolution only sets the ceiling.
- Codec efficiency sets the exchange rate. A newer codec reaches the same picture quality at a lower bitrate than an older one. That is the whole point of a new codec.
The practical version: if a video looks blocky or smeary, the fix is usually more bitrate or a better codec, not a bigger resolution number.
Video compression explained: a worked example
A salesperson records a 90 second walkaround on a phone at 1080p. The phone encodes it to H.264 in an MP4 at roughly 8 megabits per second, which lands the file at about 90 megabytes.
Sent to a platform, that file is re-encoded down to around 2 megabits per second for phone playback, so it arrives as roughly 22 megabytes and starts playing in a second or two on mobile data. On a phone screen the difference from the original is close to invisible. On a 55 inch showroom display it would not be.
Where video compression goes wrong in practice
A few failure modes turn up again and again.
- Repeated re-encoding. Every time a file is compressed again, it is compressed from an already lossy copy. Send a video through three apps and it degrades visibly at each hop.
- Messaging apps recompressing hard. A video attached to a chat message is frequently squeezed far below the quality you shot, without warning. This is why a link to a hosted page usually looks better than the file itself.
- Email attachment limits. Most mail systems cap attachments around 20 to 25 megabytes, which almost no usable video clears. Compressing until it fits is the wrong answer.
- Motion and detail. Fast pans, rain, gravel and foliage are expensive to encode. A slow, steady pan around a vehicle compresses far better than a hurried sweep, and looks better at the same file size.
One more thing worth knowing: the shape of the frame you record in affects what gets thrown away when a player has to fit it to a different shape. That side of it sits under aspect ratio.
Try Venta Capture on your own process
Build one flow for your highest-volume case type. Free, no credit card.