[Stk] Editing raw files on OS X

[Perry Cook]

Yep,

Many/most of those files date back to the 1980's, when resources were precious.

The NeXT machine had nice native sample rates of 22050 and 44100 on the 
Motorola DSP chip, along with 8k for speech.  22050 gains us two things, less
storage, and sometimes less computations.  For example, for the waveguide 
mesh or a vocal tract tube, computations are order dimension^2, and the 
dimension is linear in sampling rate so a doubling of SRATE can cost us a
4x factor on scattering computations.  Thus, 22050 was a good tradeoff that
still sounded pretty good, yet saved on storage and cycles.

It is possible that I have original 44k versions of some files, like the Mandolin
body impulse used for that commuted synthesis model.  Perhaps we'll swap
those out in a future STK release, or do away with raw files and make them
.wav, allowing you to swap your own out at will.

PRC

 
On May 16, 2013, at 11:54 AM, Gary Scavone wrote:

> That's simply the way Perry defined them when he created STK (back in the days when memory wasn't cheap).
> 
> --gary
> 
> On 2013-05-16, at 2:39 PM, Richard Dobson <richarddobson at blueyonder.co.uk> wrote:
> 
>> Ah, apparently they are 22050, not 44100. Someone else will explain why!
>> 
>> Richard Dobson
>> 
>> 
>> On 16/05/2013 19:16, Richard Dobson wrote:
>>> You can use Audacity: Import->Raw Data, use 16bit big-endian mono,
>>> 44100, no offset.
>>> 
>>> Richard Dobson
>>> 
>>> 
>> 
>> _______________________________________________
>> Stk mailing list
>> Stk at ccrma.stanford.edu
>> http://ccrma-mail.stanford.edu/mailman/listinfo/stk
> 
> 
> _______________________________________________
> Stk mailing list
> Stk at ccrma.stanford.edu
> http://ccrma-mail.stanford.edu/mailman/listinfo/stk