WAV Bit Depth Explained: 16-bit vs 24-bit vs 32-bit Float
Use 16-bit for final delivery (CD, streaming, consumer audio). Use 24-bit when recording or processing in a DAW — the extra headroom prevents rounding errors during EQ, compression, and volume automation. Use 32-bit float only for internal DAW processing or plugin chains where clipping prevention matters. You can choose all three at convertlo.pro/mp3-to-wav.html — no upload needed.
When you export a WAV file — whether from a DAW, a converter, or an audio editor — you'll be asked to choose a bit depth. Most people pick 24-bit without really knowing why, or default to 16-bit without understanding the trade-offs. This guide explains exactly what bit depth does, when it actually matters, and when a higher number is wasted space.
What Is Bit Depth?
Bit depth is the number of bits used to store each audio sample. An audio signal is a continuous waveform; digital audio converts it into a sequence of discrete measurements (samples) taken at regular intervals — that interval rate is the sample rate. Each sample value is then stored as a binary number. The number of bits available for that number is the bit depth.
Think of bit depth as vertical resolution: sample rate controls how often you measure the waveform, bit depth controls how precisely you can record each measurement's amplitude.
- 16-bit: 2¹⁶ = 65,536 possible amplitude values per sample
- 24-bit: 2²⁴ = 16,777,216 possible amplitude values
- 32-bit float: IEEE 754 floating-point — effectively unlimited precision within a huge dynamic range
Bit Depth and Dynamic Range
The primary practical effect of bit depth is dynamic range — the difference between the quietest and loudest sound the system can faithfully represent. The formula is approximately 6 dB per bit:
| Bit Depth | Quantisation Levels | Theoretical Dynamic Range | Noise Floor (approx.) |
|---|---|---|---|
| 8-bit | 256 | ~48 dB | Very audible hiss |
| 16-bit | 65,536 | ~96 dB | Below hearing threshold in quiet rooms |
| 24-bit | 16,777,216 | ~144 dB | Far below any real acoustic environment |
| 32-bit float | IEEE float | ~1,528 dB (theoretical) | Effectively zero for practical audio |
16-bit WAV — CD Quality
The CD Standard Since 1982
16-bit / 44.1 kHz is the Red Book Audio CD specification defined by Sony and Philips — the format that has been the global standard for music distribution for over 40 years. Every audio CD, iTunes purchase, and standard streaming platform delivery uses 16-bit audio.
- Dynamic range: 96 dB — more than enough for any consumer listening environment
- File size (4 min stereo, 44.1 kHz): ~40 MB
- Compatibility: Universal — every device, player, and software reads 16-bit WAV
- Best for: Final delivery, CD mastering, iTunes submission, broadcast output, consumer playback
- Not ideal when: You plan to apply heavy processing — the noise floor rises with each operation
The common misconception is that 16-bit is "low quality." In reality, 16-bit is the output format of virtually every piece of music you've ever listened to. The issue is not 16-bit for playback — it's 16-bit during processing. Every gain operation, EQ curve, or compression chain rounds sample values to the nearest integer. At 16-bit, those rounding errors accumulate. At 24-bit, they're negligible.
24-bit WAV — Studio Standard
The Professional Recording and Mixing Standard
24-bit has been the standard for professional recording, mixing, and mastering since the mid-1990s. Every modern DAW records at 24-bit by default. The difference over 16-bit isn't about what listeners hear — it's about processing headroom.
- Dynamic range: 144 dB — 48 dB more than 16-bit, far beyond any acoustic environment
- File size (4 min stereo, 44.1 kHz): ~60 MB (50% larger than 16-bit)
- Compatibility: Excellent — all modern DAWs, most modern players
- Best for: Recording sessions, mix sessions, mastering chains, DAW delivery to clients, archival
- Not ideal when: Delivering final consumer content (just adds file size with no audible benefit)
The 48 dB of extra headroom above 16-bit translates into roughly 256× more precision at the quiet end of the waveform. When you apply a -20 dB gain reduction followed by a +20 dB boost, the round-trip at 16-bit introduces rounding errors that become audible in complex sessions. At 24-bit, the errors are so small they're inaudible even after hundreds of operations.
Streaming services like Tidal Masters and Amazon Music HD distribute music at 24-bit. Apple Music Lossless uses 24-bit ALAC. If you're delivering to these platforms or to a mastering engineer, 24-bit is the correct choice for your working files — then dither to 16-bit only for standard-resolution delivery.
32-bit Float WAV — Professional DSP
Floating-Point Audio for DSP Workflows
32-bit float is a fundamentally different format from 16-bit and 24-bit PCM. Instead of fixed integers, it uses IEEE 754 floating-point representation — the same format used internally by most digital audio software. This means values can temporarily exceed 0 dBFS (full scale) without clipping, making it ideal for intermediate processing files.
- Dynamic range: ~1,528 dB (theoretical); practically, about ±770 dB of headroom
- File size (4 min stereo, 44.1 kHz): ~80 MB (2× 16-bit, ~33% larger than 24-bit)
- Compatibility: Limited — most consumer devices and players cannot play 32-bit float without conversion
- Best for: DAW internal processing, plugin chains, field recorder output (some Zoom and Sound Devices recorders output 32-bit float), post-production workflows
- Not ideal when: Delivering to end consumers, streaming, or any situation requiring broad device compatibility
The key advantage of 32-bit float is clipping protection. If a source is recorded too loud (over 0 dBFS) in 32-bit float, the waveform simply has values above 1.0 — no data is lost. You can recover a "clipped" 32-bit float recording by simply reducing the gain, because the full waveform is preserved above the digital ceiling. In fixed-point formats (16-bit, 24-bit), clipping is permanent — once samples hit the ceiling, the waveform is destroyed.
Side-by-Side Comparison
| Property | 16-bit PCM | 24-bit PCM | 32-bit Float |
|---|---|---|---|
| Format code (WAV) | PCM (1) | PCM (1) | IEEE FLOAT (3) |
| Dynamic range | ~96 dB | ~144 dB | ~1,528 dB (theoretical) |
| Quantisation levels | 65,536 | 16,777,216 | Floating point |
| File size (4 min stereo, 44.1 kHz) | ~40 MB | ~60 MB | ~80 MB |
| Values above 0 dBFS | Clips (data lost) | Clips (data lost) | Preserved (no clipping) |
| Consumer device support | Universal | Good (modern devices) | Limited |
| Streaming delivery | Standard | HD tiers (Tidal, Amazon) | Not accepted |
| DAW recording | Acceptable | Standard | Used internally |
| Processing headroom | Limited | Excellent | Unlimited |
Which Bit Depth for Your Workflow
| Workflow | Recommended | Reason |
|---|---|---|
| MP3 to WAV conversion | 16-bit | MP3 source has ≤16-bit effective dynamic range; 24-bit adds file size without benefit |
| CD mastering / iTunes delivery | 16-bit | Red Book standard; streaming re-encodes anyway |
| Recording a live session | 24-bit | Preserves headroom for post-production processing |
| Mixing and mastering in DAW | 24-bit | Processing headroom prevents accumulated rounding errors |
| Delivering to mastering engineer | 24-bit | Standard professional delivery format |
| Broadcast / EBU R128 delivery | 24-bit, 48 kHz | Broadcast spec requirement |
| Field recording (outdoor/live) | 32-bit float | Clipping protection on unpredictable sources |
| Plugin chain intermediate file | 32-bit float | Preserves headroom between processing stages |
| Streaming (Spotify, Apple Music) | 16-bit (or 24-bit for lossless tier) | Platforms encode to AAC/Vorbis regardless |
| Voice AI / Speech recognition | 16-bit, 16 kHz | Models trained on 16 kHz 16-bit PCM; higher rates waste bandwidth |
Bit Depth When Converting from MP3
A common question when using a converter like Convertlo: should I pick 16-bit, 24-bit, or 32-bit float when converting an MP3 to WAV?
The answer depends entirely on what you're doing with the output:
- Importing into a DAW for editing — 24-bit is fine but not necessary. The MP3's effective bit depth is already 12–16-bit equivalent. 24-bit just gives the DAW more room to work internally. Either 16-bit or 24-bit works.
- Submitting to a hardware sampler or game engine — 16-bit is the correct choice. Hardware samplers (Roland, Akai, Elektron) and game engines (Unity, Unreal) typically expect 16-bit WAV. 24-bit sometimes causes import errors on older hardware.
- DSP processing chain — 32-bit float is worth it here. If you're running the WAV through a chain of plugins that amplify, compress, and process, starting at 32-bit float means intermediate values never clip.
- CD mastering — 16-bit. The MP3 source limits quality regardless, and CDs are 16-bit anyway.
- Broadcast delivery — 24-bit at 48 kHz. Match the broadcast spec even when the source is MP3.
Try All Three Bit Depths — Free, No Upload
Convert any audio to 16-bit, 24-bit, or 32-bit float WAV directly in your browser. Choose from 10 sample rates (8 kHz to 192 kHz). Batch convert + ZIP download.