Re[4]: [Alsa-user] Surround Sound (5.1) How do I debug thesetup=?koi8-r?Q?=3F?=(another one)

[Sergei S. <ste...@li...>]

Lee, my point was not what JWZ said, but rather overall state of affairs
with (not only) ALSA.

I most of my "western" life I worked in VLSI - first as a design
engineer, then as simulation and verification engineer.

The latter, by the way, required even higher coding skills in a number
of languages.

I do not believe in statement by anybody regarding anything when
the statement just says "it works" or "it's supported".

I am much more willing to believe in more detailed statements like

----------------------------------------------------------------------
This card was tested by <real_person_name> for the following features:

1) stereo playback;
2) 5.1/7.1 playback (if any);
3) Line capture;
4) Microphone capture;
5) SPDIF capture/playback (do they bothe exist ?);
...

The test(s) were performed with these alsamixer/amixer settings and/or
with this asound.state file <mixer_settings_or_file_follow>.

The tests were performed with ALSA v <version_follows> with kernel v
<version_follows>.

Caveats...
Issues...
========================================================================
.

However you disagree with that, but regression testing is necessary.

If I see ALSA release 1.0.9b with the change log saying that things were
fixed to enable compilation of 2.2 and 2.4 kernels I want just to laugh
and to cry - the latter much more. It's because it speaks for itself - the
developers have not tried compilation, i.e. even basic regression tests
have not been run.

Almost the same regarding "compilation problem" bug category - I used to
be project integrator too, and the rule was that every developer was supposed
to run compilation of his block against current project tree before
giving it to me. The violations were very rare, and were not tolerated -
managers were notified immediately about such things.

And developers were supposed to run sanity tests before submitting their
modules to me.

Maybe I have a wrong mentality, but as one SW guy explained me, in VLSI
they pay much more attention to robust version control and regression testing
- new set of masks for chip manufacturing used to cost $100K in the mid-nineties,
and now it is, say $800K.

So, I "grew up" in the environment where bugs were very costly.


I am trying to say that any serious project should have some orderly
testing methodology, bug acceptance criteria - for example, muti-card
synchronization doesn't work, but most of the users have only one
card anyway, so this bug can go into a release.

I am talking about very basic things...

And I'm thinking (and maybe even dreaming) of availability of
IA32 MacOS sound drivers - someone will definitely write a binary
compatibility wrapper for them for Linux, as it has been done for
WiFi drivers for Windows - if I remember correctly.

-----Original Message-----
From: Lee Revell <rlr...@jo...>
To: Sergei Steshenko <ste...@li...>
Date: Wed, 22 Jun 2005 13:10:36 -0400
Subject: Re: Re[2]: [Alsa-user] Surround Sound (5.1) How do I debug thesetup?(another one)

> 
> On Wed, 2005-06-22 at 19:54 +0400, Sergei Steshenko wrote:
> > It's rather sad, not funny, and sad in the Owellian way.
> > 
> > Read, for example, this thread:
> > 
> > http://apple.slashdot.org/article.pl?sid=05/06/12/1315235&tid=179&tid=190&tid=1
> 
> Really, really sad.  Slashdot used to be 90% Linux developers, now it
> seems to be full of Apple fanboys who would rather complain than help
> fix the problem.
> 
> Anyway, jwz's blog post is just a bunch of sour grapes.  Allow me to
> demonstrate.
> 
> (from http://jwz.livejournal.com/494040.html)
> 
> "Remember last week, when I tried to buy _exactly the same audio card
> that 99.99% of the world owns_ and convince Linux to be able to play two
> sounds at once?"
> 
> That's the dead giveaway.  He clearly thought he was buying an SBLive!
> supported by the emu10k1 driver, with hardware mixing, which 99.9% of
> the world does own, and which has Just Worked in Linux for almost 10
> years.  But he failed to do any research and bought one of the
> snd-ca0106 "SBLive! 24 bit" devices, which is a COMPLETELY different and
> vastly inferior device.  Of course the mixing always just worked because
> IT WAS DONE IN HARDWARE.
> 
> He would have had the exact same results if he had been a Windows user
> of the kX drivers, and bought one of these devices to replace his old
> SBLive - non working sound.
> 
> If he had bothered to check the alsa-user list or even the soundcard
> matrix he would have seen that he's about the 100th Linux user to be
> duped in this way by Creative's marketing.  Taking it out on Linux is
> just juvenile.
> 
> Lee
> 
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
> from IBM. Find simple to follow Roadmaps, straightforward articles,
> informative Webcasts and more! Get everything you need to get up to
> speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
> _______________________________________________
> Alsa-user mailing list
> Als...@li...
> https://lists.sourceforge.net/lists/listinfo/alsa-user
>