Hi,
Might be of interest to some people here; Pizza correctly pointed out that
this should go to the list.
llap,
Christoph
---------- Forwarded message ----------
Date: Thu, 31 Jan 2002 22:48:21 -0500
From: Stuffed Crust <[EMAIL PROTECTED]>
To: Christoph Reichenbach <[EMAIL PROTECTED]>
Subject: Re: PCM layer abstraction
On Fri, Feb 01, 2002 at 12:32:37AM +0100, Christoph Reichenbach wrote:
> has the PCM layer been abstracted sufficiently? We're currently
> considering a OS X coreaudio backend for FreeSCI, because the thing used
> by SDL doesn't survive fork(), and we'll need that until we have migrated
> to a better sound architecture that does not depend on the sound server as
> a separate process or thread.
The PCM output layer is as about as simple as you can get. There are
public open/close calls, and it's up to the actual driver to figure out
the timing/callback/whatever. There's a fill_buffer call to the upper
layer of the PCM subsystem, but that's it.
The middle of the PCM subsystem needs a whole lot of work, but that's
completely independent of the actual output drivers.
So yes, writing another pcmout driver should be trivial at this point.
The only caveat is that it needs to be 48000Hz/16-bit/mono, because
being able to configure that would involve a lot more than the pcmout
layer (and I decided to put that off until it was genuinely needed)
Any reason why this wasn't cc:ed to the list? (feel free to)
I'll try to pop into #freesci tomorrow.
(Work's been keeping me quite busy; it's also music code in a manner of
speaking. :))
- Pizza
--
Solomon Peachy pizzaATfucktheusers.org
I ain't broke, but I'm badly bent. ICQ# 1318344
Patience comes to those who wait.
...It's not "Beanbag Love", it's a "Transanimate Relationship"...
-- Attached file included as plaintext by Listar --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8WhAFysXuytMhc5ERAjElAJ4vhkYXh+GeWnjxkOAQzra+hWYVfwCggK6p
y84onLZAQFCcuhA/wHjofuk=
=2yoM
-----END PGP SIGNATURE-----
