Re: [alsa-devel] Re: [linux-audio-dev] Re: Toward a modularizationofaudio component

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: Re: [alsa-devel] Re: [linux-audio-dev] Re: Toward a modularizationofaudio component
From: Paul Davis (pbd_AT_Op.Net)
Date: Tue May 08 2001 - 00:55:46 EEST


>> >An audio component may treat only *one* format, nothing force it to
>> >handle all.
>>
>> In your model Abramo, what will happen if someone writes a 16bit,
>> interleaved stereo component and tries to use it with an engine that
>> uses 32bit non-interleaved float? Will there be an error, or just
>> silent data format conversion?
>
>Error, of course (on params negotiation).

bwahahahahaha! :))

"params negotiation"? you're suggesting that every component in this
audio system uses the ALSA parameter negotiation system? pardon me,
but the *entire* point of a system like LAAGA is abstraction. the ALSA
params model is *necessary* when dealing with audio hardware. its a
complete monstrosity when dealing with software plugins. in the
proposed API's for LAAGA, there are no parameters except for the
engine passing the maximum number of frames it will pass to the
process() callback of a plugin. plugins can't alter it, so its not
really a parameter in the conventional sense.

and there's a practical point that if anyone tries to alter the period
size, the requirement/guarantee of perfect synchronicity seems rather
hard to meet. you could only do that by enforcing a single period
size, and then there's not much point in having an API that lets you
set, is there?

Abramo, why do you think ASIO and VST (not to mention ReWire) exist as
totally separate API's? do you think that steinberg have missed the
obvious, or that they lack imagination, or that they are malicious or
what?

--p


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Tue May 08 2001 - 01:15:43 EEST