192khz .. playback fidelity lower than 44.1khz?

Tony Hardie-Bick tony at entity.net
Tue Mar 6 17:19:28 CET 2012


On 06/03/12 15:29, Andrew Robinson wrote:
> I was talking about this with a studio owner a few years back. After a
> heavy session on A/B testing the first few 24/96 converters on the
> market, he'd come to the opinion that after 20 years of practice, the
> industry had got very good at making good sounding 16/44.1 kit, and
> that the 24/96 stuff needed a good few years to get up to speed. He
> could definitely tell the difference, but he didn't like it, comparing
> it to the awful high-end heavy mastering on the earliest CDs.

When processing audio samples and doing something with them (sample rate 
conversion for example, or band-limited synthesis, or filtering) a much higher 
sample rate can make it easier to compute the result.

This is why stomp boxes running at 96kHz generally sound better than those 
running at 48kHz.

Also, 24 bits is way more than necessary for playback, but during recording, one 
doesn't know in advance what the maximum signal level will be, so, you can hope 
to be lucky, or tolerate some overloads, or play safe and deal with the 
compromise in bit depth. In the latter situation, the more precision your AD 
gives you (true precision, not just the number of bits it chucks out), the more 
thankful you're gonna be at the end of the recording session.

As far as playback is concerned, it's true that the presence of ultrasonics can 
be a problem. Looking at the data sheets for AD and DA converters running at 
these frequencies, there are suprisingly relaxed parameters for pre and post 
conversion filtering, so IMO this is likely a significant cause of the problem.

Tony (HB)


More information about the music-bar mailing list