[linux-audio-user] Sample Rates / Sample Sizes

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: [linux-audio-user] Sample Rates / Sample Sizes
From: davidrclark_AT_earthlink.net
Date: Sat Jun 19 2004 - 18:03:56 EEST


Starting a new thread originating as:

  Re: producing a drum sample library for hydrogen

I can easily tell the difference, as can many, between 16- and 24-bit
samples, but not equally for all instruments, of course. I highly, highly
recommend using 24-bit samples for any widely used library, especially
given the amount of processing that most of us do. I myself also use
96,000 samples/second because I always do quite a bit of processing, but
that's a lot tougher sell for a library.

---------------------

Someone said that 44,100 samples/second is a quirk. It's not merely a
quirk. 44,100 = 2^2 * 3^2 * 5^2 * 7^2 which has some useful properties
compared to 48,000 = 2^7 * 3 * 5^3. Lots of small primes that differ is
sometimes better than a bunch of powers of two, despite the advantage
primarily associated with FFT's that some focus on. Looked at from a
prime number standpoint (therefore certain types of algorithms),
48,000 is truly bizarre compared to 44,100. 96,000 is yet ANOTHER power
of two on top of 48,000! Whether a particular sample rate is a good one
or not is dependent upon more than just what is customary and standard.
 


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Sat Jun 19 2004 - 17:56:24 EEST