Use the free tool below: your data never leaves your browser — all work happens on your device, and nothing is uploaded to any server.
What bit depth actually controls
Bit depth is the number of bits used to describe the loudness of each audio sample. Think of it as the resolution of a ruler: with more marks on the ruler, you can measure smaller differences. A 16-bit recording has 65,536 possible loudness levels for every sample, while a 24-bit recording has more than 16 million. That sounds like a huge difference, and in numbers it is, but what you hear is measured in dynamic range: the gap between the loudest sound the system can hold and the quietest detail it can still describe. The calculator above turns any bit depth into that range with the standard formula, 6.02 times the bit depth plus 1.76, and it also works out the uncompressed file size for a recording of a given length.
The formula is worth understanding, because it explains why each extra bit is worth about 6 decibels. Every time you add one bit, you double the number of levels, and doubling a signal’s resolution is worth roughly 6 decibels of range. So 8-bit audio gives about 50 decibels, which is why old game consoles and telephone-style sounds feel noisy and harsh: quiet details fall below the noise floor and turn into audible hiss and distortion. 16-bit audio, the CD standard, gives about 98 decibels, which comfortably covers most music, from a whispered vocal to a full drum kit, as long as the recording level is set sensibly.
How to use the calculator
- Pick the bit depth you are considering, from 8 to 32 bits.
- Choose the sample rate and the channel count, mono or stereo.
- Enter how long the recording will last, in minutes.
- Read the theoretical dynamic range, the number of quantisation levels, and the uncompressed file size.
A worked example
Take a 5-minute stereo recording at 44,100 Hz and 16 bits. The calculator gives a dynamic range of 98.08 decibels and 65,536 levels. The file size is about 50 megabytes uncompressed, because 44,100 samples per second times 2 bytes per sample times 2 channels times 300 seconds is roughly 52.9 million bytes. Switch the same recording to 24 bits and the range rises to 146.24 decibels in theory, while the file grows by half, to about 76 megabytes. The extra bits do not make loud sounds louder; they make quiet sounds cleaner, because the noise floor drops further below the music.
That last point is the one most people get backwards. Beginners often record as hot as possible, pushing levels close to clipping, because they fear losing quality. With 24-bit recording, the opposite approach works better: leave generous headroom, peak around minus 12 decibels, and let the extra bits protect the quiet passages. The noise floor at 24 bits is so low that even a recording peaked at minus 18 decibels still holds far more clean detail than a 16-bit recording pushed to the edge. Clipping, once it happens, cannot be undone, while a slightly quiet 24-bit recording can be turned up cleanly later.
Why real gear never reaches the theoretical number
The calculator reports the theoretical range, and honest manufacturers will tell you their converters fall short of it. Analogue electronics add noise before the signal ever reaches the digital stage, so a real 24-bit interface typically delivers 110 to 120 decibels of usable range, not 146. That is still excellent, and far beyond what most rooms and microphones can use, since a quiet bedroom already has 30 decibels or more of background noise from fans, traffic and plumbing. Chasing the last few theoretical decibels matters far less than treating the room, placing the microphone well, and setting sane levels.
File size is the practical cost of higher bit depths. A full day of 24-bit multitrack recording can fill a drive quickly, and streaming services compress everything anyway, so delivering a finished mix at 24 bits is about preserving quality through mastering, not about what listeners hear on their phones. For most home recording, 24-bit capture and 16-bit or compressed delivery is the sensible split: capture with room to spare, deliver in the format the audience actually uses.
Common mistakes
- Recording too hot at 24 bits out of habit from the 16-bit era, and clipping takes that cannot be repaired.
- Assuming 32-bit float recording fixes everything: it forgives level mistakes, but it cannot fix a bad room, a wrong microphone position, or distortion that happened before conversion.
- Judging bit depth by file size alone: a larger file is not automatically a better recording if the source was noisy.
- Forgetting dither when reducing bit depth: going from 24 bits to 16 bits for delivery needs dither, a tiny amount of shaped noise that prevents harsh quantisation distortion in quiet fades.
Who uses this calculation
Home studio owners use it when choosing interface settings, podcasters use it to understand why their quiet guests sound noisy, and students use it to connect the abstract idea of bits to something audible. Run your own numbers, then listen: record the same quiet passage at 16 and 24 bits, turn both up equally, and hear where the noise floor sits. The calculator gives you the theory; your ears give you the reason it matters.