[MPlayer-dev-eng] [patch] prefer ALSA over OSS

[Vladimir Mosgalin]

Hi Rich Felker!

 On 2007.05.04 at 20:57:33 -0400, Rich Felker wrote next:

> > Well this is completely OT. The original problem is that it doesn't
> > happen with oss interface in alsa, because modern hardware in a way is
> > even more limited than old one, despite being better. The lack of
> > hardware mixing and random sample rate/sample format support in hardware
> > is a very good thing, quality-wise.
> 
> No, it's a bad thing for quality. No degree of digital signal
> processing can give the correct result you'd get from correctly timing
> the output clock to the samplerate of your recording; any resampling

You are wrong. There are _no_ problems from good digital resampling in
high resolution (move to 24 bits resolution to save original original
16-bit precision - and of course all audio processing must be at least
24-bit, most likely float though).

However, there could a lot of problems with jitter and other negative
effects on DAC if you clock isn't fixed.

As a matter or fact, most DACs resample digital input once more, to some
higher frequency, like 96khz or 192khz and it makes quality better.
Because of two reasons: first, it doesn't make original digital stream
any worse and second, the DAC unit by itself is designed such way
so it produces better analog output when feeded by hi-resolution signal.

The problem in your point is that you consider only digital stream.
However, the real source of problems is DAC, a moment where conversion
to analog takes place. It's a very.. how should I say.. fragile unit,
and effects that aren't even considered in pure digital world can
destroy it.

> involves tradeoffs between artifacts, destruction of high frequency
> content, etc.

Then stop using some kind of very broken resampler and try real one.

> > Because it can be done in software and therefore should be done in
> > software,
> And just a minute ago you were saying doing things in software (3d
> graphics) was a bad idea.. Now you say resampling should be done in
> software just because it can? Make up your mind!

[went in separate letter]

> > > In the fastest mode, it's slower than libavcodec's resampler.
> > 
> > No one cares. In normal mode, it takes marginal amounts of cpu.
> And sounds like utter shit.

Why do you care, if you can't give any sample that proves it anyway and
haven't even said that you heard it at least once?

> > You cannot prove that. I believe what my ears tell me: its quality is
> > good. I don't want to discuss libavcodec now, but if quality is already
> > _good enough_, if it is _very good_, why bother saying that "it's lower
> > than something", if you ears won't notice that anyway?
> Do listening tests. It was done by someone I know and found to be very

I did ;) And I'm extremely satisfied with its quality.

> bad. Also, max quality mode in ALSA is so slow it's unusable, so it's
> irrelevant.

I find it acceptable cpu-wise, but it has worse quality that normal
mode, so just forget about it.

-- 

Vladimir