Yes and no. Yes, it's "impressive" that you can work with a wide range of values. You have a huge working space. But no, the extra "dynamic range" is not suitable for microphone data acquisition because it's inherently imprecise. It's only useful for computer generated data. DEFINITION: Audio fidelity refers to how accurately a copy of audio reproduces its original source. If you have inaccurate numbers how can you accurately copy the original source? You can't. As I mentioned, microphones don't need more than 12-bits, so 24-bit is overkill, let alone 32-bit float. BUT IF microphones could do more than 24-bit, BELIEVE YOU ME, everyone would use 32-bit FIXED! :)
so what you are saying is with 32 bit float recording, you get at least 24bit quality plus the extra dynamic range ?
that's still impressive
Yes and no. Yes, it's "impressive" that you can work with a wide range of values. You have a huge working space. But no, the extra "dynamic range" is not suitable for microphone data acquisition because it's inherently imprecise. It's only useful for computer generated data.
DEFINITION: Audio fidelity refers to how accurately a copy of audio reproduces its original source. If you have inaccurate numbers how can you accurately copy the original source? You can't.
As I mentioned, microphones don't need more than 12-bits, so 24-bit is overkill, let alone 32-bit float. BUT IF microphones could do more than 24-bit, BELIEVE YOU ME, everyone would use 32-bit FIXED! :)