On the ALSA Track (Linux Journal) [LWN.net]

The Linux Journal has posted an introduction to the ALSA sound system by Dave Phillips. "Like Linux itself, ALSA began with rather modest goals: Jaroslav [Kysela] simply wanted more out of his Gravis UltraSound soundcard than the existing API could deliver, and he was willing and able to meet the demands of the task. Like Linus Torvalds, Jaroslav eventually found himself at the center of a group of talented developers, all dedicated to the development of a superior audio API for Linux."

(Log in to post comments)

On the ALSA Track (Linux Journal)

Posted Feb 12, 2004 7:59 UTC (Thu) by zettberlin (guest, #19389) [Link]

It is strange, that ALSA need to be propagated as if it would be something new - i use ALSA in SUSE and MDK for 2 Years. OSS is in Fact completely depractated.
I cannot understand, that RH fails to implement ALSA - this is the main Reason for me , to stop trying RH - no usable PacketmanagerGUI, no MP3, no ALSA.

On the ALSA Track (Linux Journal)

Posted Feb 12, 2004 20:09 UTC (Thu) by nkoozer (guest, #5553) [Link]

RH has functional sound where many others do not. The complexity of ALSA exceeds the means of getting this complexity to function on the plethora of machines in the wild.

On ALSA being "new"

Posted Feb 13, 2004 15:31 UTC (Fri) by Duncan (guest, #6647) [Link]

Keep in mind that the 2.4 kernel on which most current distribs are based
defaults to OSS. ALSA has been patch to the official kernel, until the 2.5/2.6
series. While many distribs (including your SuSE and my Mandrake) have
often included both and based the decision on what system and module to use
on what is supposed to work best on the detected hardware, the fact remains
that ALSA has been an add-on patch to the mainline stable kernel, until 2.6,
which just came out.

Based on that, there are probably quite a number of folks that have never
used ALSA and are (or will be) getting their first taste of it with the 2.6
kernel. Many in fact will not load their first 2.6 kernel until the distribs they
use come with it, and only THEN will they see ALSA.

Thus, the new ALSA focus is reasonable, given that it is just part of the focus
on the new kernel and what's different about it. It's just part of the general
hubbub and publicity surrounding the new kernel, is all, and as such, it's to be
expected. Nothing more, nothing less, really, tho I imagine those who've
worked hard on it are proud to see it finally coming into its own, full kernel
recognition and all. I know I'd certainly be happy and proud of "my baby
growing up" like that.

Duncan

Score only 4/5 for ALSA's aims

Posted Feb 13, 2004 15:50 UTC (Fri) by Wills (guest, #1813) [Link]

The problem with saying "OSS is in fact completely deprecated" is that only four out of the five aims of the ALSA project have been implemented. I am referring to aim #5 ("Support for the older OSS API") which is not yet fully implemented. Until aim #5 is completed, ALSA cannot be a valid candidate for completely replacing OSS and OSS remains essential. The fact is there are major bugs in ALSA's OSS-mode sequencer which is seriously broken (http://bugtrack.alsa-project.org/alsa-bug/bug_view_advanced_page.php?bug_id=28). The bug report explains how to reproduce these bugs.

Fortunately I think these are the only major bugs that prevent ALSA being a compatible replacement for OSS. However, it has been more than one year since the bugs were first reported to the alsa-devel mailing list (26 Aug 2002) and nobody has committed any patches to ALSA CVS or to the mailing list. The bugs are in the latest ALSA CVS and have been in ALSA since at least ALSA 0.3.0 (10 Feb 1999).

I wish I knew the ALSA driver code well enough to be able to fix the bugs myself. I appreciate the work of the ALSA project and hope somebody will get around to fixing these bugs soon.

Will