What CBR and VBR describe
Constant bitrate (CBR) aims to use a fixed target rate over time. Variable bitrate (VBR) lets the encoder allocate bits differently as the signal changes and is often selected through a quality level or an average target. These terms describe bitrate mode, not the codec: MP3, AAC, and other codecs may expose different mode controls.
The calculator deliberately accepts an average VBR bitrate because file-size arithmetic needs an average. It does not include a “complexity” slider that claims to know the encoder result from a label. Only a real encode can measure the final bytes.
Interactive CBR vs VBR file-size formula
The center formula is bytes ≈ durationSeconds × averageBitrateKbps × 1000 ÷ 8. The optional allowance is added transparently for container and metadata. For VBR, the uncertainty percentage expands the center into a range; it does not assert that a particular song or speech recording will land at either boundary.
The comparison reports CBR bytes, VBR center bytes, the VBR range, and the signed center difference. If both average bitrates are identical, their payload centers are identical. VBR is not mathematically guaranteed to be smaller.
Why VBR is not automatically smaller or better
At the same nominal average, VBR may distribute bits more selectively, but the outcome depends on codec, encoder implementation, selected mode, source, and constraints. A high VBR target can be larger than a low CBR target. A poor encode can remain poor regardless of the mode label, and a larger number is not proof of transparent sound.
Listening should use actual files at matched conditions. The Audio Bitrate Comparison can generate measured versions serially for personal evaluation without assigning a quality percentage.
Compatibility and streaming decisions
Compatibility requirements sometimes favor predictable rate behavior, but modern playback support varies across codec, container, decoder, and delivery system. Avoid turning one host’s current recommendation into a universal rule. Validate the actual publishing, streaming, editing, or embedded environment.
CBR can be useful when a simple fixed throughput estimate or legacy workflow matters. VBR can be useful when an encoder’s quality-oriented allocation is appropriate. Neither choice replaces testing the generated container and codec.
Container, tags and encoder differences
Real files contain headers, seek tables, padding, tags, and sometimes artwork. An encoder may also miss the target average slightly or apply reservoir and framing rules. Those bytes explain why a file can differ from the payload formula even when duration and reported average bitrate look correct.
For short audio, fixed overhead can be a visible percentage. For long audio, the bitrate payload normally dominates. Use actual output size for a hard upload limit and keep a safety allowance in any target calculation.
Decision framework for common audio work
| Use case | Question to ask | Practical direction |
|---|
| Speech delivery | Is broad playback or smallest acceptable copy more important? | Test codec and a modest average; mono may matter more than mode |
| Streaming compatibility | Does the real delivery system specify a supported mode? | Follow the verified system, not a generic claim |
| Archive | Must all source information remain? | Keep PCM or lossless; CBR/VBR is a delivery choice |
| Music sharing | Have you compared actual encoder outputs? | Listen and measure instead of trusting the mode name |
Related parameters
Sample rate describes time-domain sampling, while bitrate describes encoded data per second. Read Sample Rate vs Bitrate to avoid multiplying sample rate into a fixed-bitrate estimate. Read Lossy vs Lossless Audio when the more important decision is whether a delivery codec may discard information.
For a real episode workflow, the Podcast Audio Compressor combines audited presets with measured output and publishing-storage planning.