32-bit float vs 24-bit for Recording.
Any advantages or disadvantages to setting DAW to 32-bit float other than the increase in files size? Does 32 help prevent overloading and distortion while recording?
32-bit float vs 24-bit for Recording
-
- KVRist
- 108 posts since 30 May, 2019
If you dont have an audio interface that can record at 32 bit float then use 24 bit.
The obvious advantage to 32 bit float is that you can record at any level and not worry about digital clipping.
The obvious advantage to 32 bit float is that you can record at any level and not worry about digital clipping.
-
- KVRAF
- 16806 posts since 8 Mar, 2005 from Utrecht, Holland
Some considerations:
* 24bit int is sometimes stored in 32bit words, because that is more efficient in terms of speed. Shoving four 3-byte samples into three 4-byte words is an extra computation. You need to test it per DAW, do some simple calculations to check it.
* A 32bit floating point number is technically a 24bit integer with a 8bit exponent. To increase the range above the upper limit and below near-zero.
* In the range used for audio recordings, a 20bit integer is enough. No A/D or D/A converter goes beyond a dynamical range of 120dB, which is about the limit of what's possible in analog. (each bit gives 6dB dyn.range)
* imho even 16bit suffices if you have proper control over your levels aka The Art of Gain Staging. This can represent a dynamical range of 96dB. My mixer is specced at 89dB, so...
* All DAWs will first convert your recorded audio to 32bit floats because that is the format all VST effects use. Digital clipping might only happen at the very last point: when it hits the D/A converter or you export to WAV.
* 24bit int is sometimes stored in 32bit words, because that is more efficient in terms of speed. Shoving four 3-byte samples into three 4-byte words is an extra computation. You need to test it per DAW, do some simple calculations to check it.
* A 32bit floating point number is technically a 24bit integer with a 8bit exponent. To increase the range above the upper limit and below near-zero.
* In the range used for audio recordings, a 20bit integer is enough. No A/D or D/A converter goes beyond a dynamical range of 120dB, which is about the limit of what's possible in analog. (each bit gives 6dB dyn.range)
* imho even 16bit suffices if you have proper control over your levels aka The Art of Gain Staging. This can represent a dynamical range of 96dB. My mixer is specced at 89dB, so...
* All DAWs will first convert your recorded audio to 32bit floats because that is the format all VST effects use. Digital clipping might only happen at the very last point: when it hits the D/A converter or you export to WAV.
We are the KVR collective. Resistance is futile. You will be assimilated. 
My MusicCalc is served over https!!
My MusicCalc is served over https!!