Video File Size Calculator
Estimate a video file from exact duration, average video bitrate, audio bitrate and track count, timed-data bitrate, and a user-entered container overhead. Then plan delivery copies, storage-card runtime, and idealized upload time without confusing bits with bytes or GB with GiB.
This is an average-bitrate estimate. Variable bitrate, codec behavior, muxer implementation, edit structure, metadata, thumbnails, filesystem allocation, encryption, cloud billing, platform transcodes, proxies, renders, and caches can make real storage different.
Enter one media asset
Program duration
Average stream budget
Storage and delivery
Estimated asset footprint
Binary-unit view will appear here.
Timecode and per-minute storage will appear here.
—
—
—
—
Bitrate-to-file-size calculation
Bitrate is bits per second. File size is commonly displayed in bytes, and NIST identifies one byte as eight bits. The calculator converts the video rate from megabits per second and each audio/data rate from kilobits per second using decimal SI prefixes, adds all streams, multiplies by duration, and divides by eight.
container bytes = payload bytes × (1 + overhead %)
The default picture stream is 25 Mbps. Two audio tracks at 320 kbps contribute 640 kbps, and captions/data add 50 kbps, so the combined payload rate is 25.69 Mbps. Across 5,400 seconds, payload is 17.341 decimal GB. A three percent entered overhead adds about 0.520 GB, producing an estimated 17.861 GB file.
Overhead is not a universal percentage. Container indexes, sample tables, edit lists, subtitles, timecode, metadata, attachments, thumbnails, padding, fragmentation, and error-recovery features vary. For a known muxer and format, measure several representative outputs and replace the example allowance with an observed planning value.
Decimal MB/GB and binary MiB/GiB
NIST states that SI prefixes are powers of ten: one MB is 1,000,000 bytes and one GB is 1,000,000,000 bytes. Binary prefixes are different: one MiB is 1,048,576 bytes and one GiB is 1,073,741,824 bytes.
The same file therefore has a larger numerical count in decimal GB than binary GiB. The default 17.861 GB file is about 16.634 GiB. Neither is a different file; only the display unit changes. Storage vendors commonly state decimal capacity, while an operating system or application may show binary capacity under a “GB” label.
Keep Mbps separate from MB/s
A 25 Mbps stream is not 25 megabytes per second. Divide bits by eight: 25 Mbps alone is 3.125 MB/s before audio, data, and overhead. Uppercase B conventionally means byte; lowercase b means bit.
Network speed tests usually report decimal Mbps. This calculator uses entered Mbps as ideal payload throughput and does not subtract TCP/TLS, retransmission, Wi-Fi, VPN, cloud-ingest, or server overhead. Real transfer takes longer and can vary over time.
Choose an average bitrate that belongs to the workflow
| Input | What belongs there | Frequent mistake |
|---|---|---|
| Video Mbps | Average compressed video bitrate for this exact codec, resolution, frame rate, bit depth, chroma, profile, and content. | Copying a camera sensor rate or platform maximum that does not describe the exported file. |
| Audio kbps per track | Encoded bitrate for one mono, stereo, or multichannel stream, multiplied by actual stream count. | Calling channels “tracks” or forgetting alternate language and descriptive audio. |
| Timed data kbps | A measured or budgeted average for subtitles, captions, telemetry, and timed metadata. | Assuming attached artwork or a large metadata block is a constant stream. |
| Overhead percent | A workflow-specific allowance derived from test muxes when possible. | Using it to hide uncertainty in video bitrate. |
Resolution alone cannot determine size. A complex grainy scene, fast motion, noise, and frequent cuts may require more bits than a clean static interview at the same dimensions. Codec efficiency and encoder settings matter. Constant-quality encoding creates whatever average bitrate the content requires; estimate it with a representative test, not a generic resolution chart.
Plan the storage beyond one master
The copies field multiplies only the final estimated container. A resilient production plan may also need camera originals, separate audio, project files, graphics, fonts and licenses, proxy media, optimized media, render cache, conform media, review exports, captions, checksums, parity, backups, and archive replicas. Those are different assets and should be listed separately.
Nominal card runtime divides entered decimal capacity by estimated bytes per second. It assumes an empty card makes every advertised byte available. Filesystem formatting, reserved space, maximum file size, camera spanning behavior, write-speed headroom, bad-block management, and a safety reserve reduce practical capacity. Recording reliability depends on sustained write speed and camera-approved media, not only total GB.
For delivery, compare checksum-verified copies rather than assuming a completed progress bar proves integrity. Record filename, size in bytes, duration, codec/profile, frame rate, resolution, audio layout, color metadata, caption format, checksum algorithm and value, and transfer date. A byte-exact manifest makes later reconciliation much easier.
Variable bitrate and test-encode audit
For VBR, the entered bitrate must be the expected average across the full program. A “maximum bitrate” predicts an upper-bound stream budget only if the encoder sustains that maximum, which it usually does not. A target bitrate may also differ from the achieved average because of rate-control limits, short duration, scene complexity, multipass analysis, or encoder implementation.
Encode a representative segment containing difficult motion, texture, gradients, noise, titles, captions, and all audio layouts. Inspect technical quality and measure actual file bytes and duration. Actual average container bitrate equals file bytes × 8 ÷ seconds. Use several segments or a full draft when the program varies substantially.
Do not scale a ten-second sample blindly if it contains proportionally large headers or is not representative. Compare payload and container behavior, then keep an explicit contingency. For regulated, contracted, or platform delivery, the published specification and validation report control over this estimate.
Build separate manifests for each deliverable
A mezzanine master, streaming upload, social cut, review copy, archive package, and cinema or broadcast deliverable should not share one bitrate assumption. Create a row for each asset with its own duration, picture codec, rate control, audio layout, captions, metadata, container, and required copies. Summing distinct results is more defensible than multiplying one “average video” by the number of versions.
Language and accessibility versions need careful track accounting. A master with one 5.1 mix, one stereo fold-down, two alternate languages, and audio description may contain several audio streams with different channel layouts and bitrates. Enter the bitrate for one stream only when all streams match; otherwise calculate their combined kbps separately and place that sum into one track for the size estimate while preserving the detailed manifest elsewhere.
Frame rate does not appear as a direct input because its storage effect is already embodied in the encoded average bitrate. Uncompressed and lightly compressed production formats are often specified by frame dimensions, frame rate, chroma sampling, bit depth, and packing; for those, first derive or obtain the actual data rate from the format specification. Do not enter a guessed compressed rate merely because two files share resolution.
When a delivery limit is strict, reverse the arithmetic. Reserve known audio, data, and overhead; divide the remaining byte budget by duration; then derive a maximum average video bitrate. Leave margin for rate-control overshoot and validate the complete muxed output. This calculator currently runs in the forward direction and does not enforce a maximum file size.
Video file size questions
Can bitrate predict the exact file size?
Only when the actual average total bitrate and container structure are known. Before encoding, bitrate is an assumption and overhead varies. Use the estimate for capacity planning, then verify with a representative or complete export.
Which bitrate should I enter for VBR?
Enter the expected achieved average video bitrate, not only the peak or maximum. If the encoder uses constant quality without a dependable average target, measure a representative test and treat the result as a scenario.
Are audio channels the same as tracks?
No. One encoded track can contain mono, stereo, 5.1, or another channel layout. Enter bitrate for one encoded audio stream and count the actual streams, such as main mix, alternate language, and descriptive audio.
Why does my computer show fewer GB?
The vendor may state decimal GB while software displays binary GiB or reserves filesystem capacity. The calculator shows both. Also confirm whether the application rounds, includes hidden data, or reports allocated rather than logical size.
Why is the real upload slower?
Entered Mbps is treated as continuous payload throughput. Network protocol overhead, retransmissions, congestion, Wi-Fi, routing, VPN, storage reads, encryption, and server ingest can all add time. Measure sustained upload speed to the actual destination.
Can I use the card runtime to choose recording media?
Capacity is only one requirement. The camera may require approved media, sustained write speed, format, filesystem, file-spanning support, and safety headroom. Follow camera and media manufacturer specifications and test the full recording mode.
References
- NIST — Definitions of the SI Units: Binary Prefixes. One-byte/eight-bit relationship and MB/MiB, GB/GiB distinctions.
- NIST Special Publication 330, Section 3. Decimal SI prefixes and separate binary prefixes.
- NIST — Metric (SI) Prefixes. Current recognized decimal prefix factors and symbols.
- NIST Glossary — Mbyte. Practical ambiguity in “megabyte” usage and the need to state the basis.