Subject: RE: [linux-audio-dev] an open letter to Linus re: low latency
From: Garth Brantley (garth_AT_fullduplex.com)
Date: Fri Jun 23 2000 - 16:59:28 EEST
> -----Original Message-----
> From: Paul Barton-Davis [mailto:pbd_AT_Op.Net]
> Sent: Thursday, June 22, 2000 10:35 PM
> To: linux-audio-dev_AT_ginette.musique.umontreal.ca
> Subject: Re: [linux-audio-dev] an open letter to Linus re:
> low latency
>
...
>
> This distinction between "close to the h/w" and "i want my API" is
> even more true for audio. There is no way in hell that you would
> sensibly do real-time FX on a networked audio connection. For such
> things, you want to sit nice and close to the audio interface. But for
> many things (WM audio messages, for example), you really want a nice
> API that handles networking transparently.
>
I think the really big underlying problem here is that neither the
architecture of the personal computer or current network topology are well
suited to handling realtime data/proccessing. It occurrs to me that the
determining decisions were made long before Linus's time. Maybe instead of
adopting audio I/O cards and standardizing the ASIO interface for all this
time, we should have been standardizing a DSP card architecture & software
tools to develop for it. Just a thought.
-Garth
> As a result of this distinction we should not expect a single uniform
> "audio subsystem" that all audio apps could use. The same thing is now
> happening with X, with the emergence of DRI and Berlin, except that
> its happening after X's widespread adoption.
>
...
> --p
This archive was generated by hypermail 2b28 : Fri Jun 23 2000 - 17:33:34 EEST