How Video File Size Is Calculated

File size in video is not mysterious. It is bitrate multiplied by duration, converted from bits to bytes. Everything else is detail around the edges of that one sum.

The formula

Size in bytes equals total bitrate in kbps, times 1,000, times duration in seconds, divided by 8. The total bitrate is the video and audio added together, because both are stored.

A worked example: five minutes at 5,000 kbps video and 128 kbps audio. That is 5,128 kbps over 300 seconds, or 1,538,400,000 bits. Divided by eight, 192,300,000 bytes. A file manager will show that as about 183MB.

Why the number on screen differs

That last step is where confusion lives. Bitrates use decimal prefixes: a kilobit is exactly 1,000 bits. File sizes usually use binary ones: a megabyte shown by an operating system is normally 1,048,576 bytes. The two conventions disagree by about 4.9 per cent.

This is why a file calculated at 25MB can appear as 23.8MB, and why a video exported to fit a limit sometimes misses it by a hair. There is a formally correct vocabulary, in which the binary units are called mebibytes and gibibytes, but almost nobody uses it outside standards documents. The practical response is to leave a few per cent of headroom.

What resolution has to do with it

Resolution does not appear in the formula at all, which surprises people. It affects size only through the bitrate you choose: more pixels need more bits to look acceptable. A 4K file encoded at the same bitrate as a 1080p one is exactly the same size and looks considerably worse.

The same is true of frame rate. Sixty frames a second does not double the file by itself; it means each frame gets half as many bits at the same bitrate, so you generally raise the bitrate to compensate, and that is what makes the file bigger.

Container overhead

The finished file also carries an index of where every frame sits, plus metadata and any subtitles or chapters. For a normal MP4 this adds roughly half a per cent. Negligible on a short clip, a few megabytes on a long one.

When the size cannot be predicted

All of the above assumes the encoder is targeting an average bitrate. When it targets a quality level instead, it decides the bitrate scene by scene and the final size is genuinely unknown until it finishes. Complex footage produces a larger file than simple footage at the same setting, and that is the mode working as intended.

Two tools on this site do these sums: the file size calculator goes forwards from bitrate to size, and the bitrate calculator goes backwards from a size limit to the bitrate that fits.