Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 02 Nov 1998 14:18:38 -0800
From:      grady@xcf.berkeley.edu (Steven Grady)
To:        multimedia@FreeBSD.ORG
Subject:   Re: How can we switch to a higher-level audio interface? 
Message-ID:  <199811022217.OAA15557@hub.freebsd.org>
In-Reply-To: Your message of Mon, 2 Nov 1998 21:03:53 %2B0100 (MET)  <199811022003.VAA14984@labinfo.iet.unipi.it> 

next in thread | previous in thread | raw e-mail | index | archive | help
> there is one big problem with audio: unless video, you want real time
> response and a userspace server cannot always help you, let alone the
> jitter and losses you can have with a networked audio server.

It's true you are not guaranteed a real-time response (although FreeBSD
does have real-time scheduling, or at least some indications, eg.  the
man page for rtprio; I don't know anything about it).  But a
well-written server is not likely to have jitter or losses -- it should
certainly be able to keep the buffer filled; and if it can't, then an
application program certainly couldn't do any better.  So I don't think
that's an excuse.

> Also, there is no equivalent of the "current window"/"input focus" for
> audio: all output is mixed together.

Agreed, but to me this is an interesting design problem, not an
insurmountable obstacle.  For instance, perhaps you need a "source
manager" instead of a "window manager".  In X, you can work without a
window manager, but it will be painful.  The source manager could allow
you to select one of n source audio streams as the foreground, and have
that one have the highest relative volume.  If you don't have one
running, everything has the same volume.

This idea is off the top of my head; I'm sure people could come up with
better solutions.  Again, the point is that there are indeed
interesting problems (some of which the people who've worked on NAS and
other technologies have already thought about), but they do not mean
that no good solution is possible.

(To take the simplest case -- Windows is a proof by example -- at least
it's possible to have multiple simultaneous sounds.  I sure hope we
could do better than Microsoft...)

	Steven

"`Simpson', eh?  I'll remember that name!"

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-multimedia" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199811022217.OAA15557>