Subject: Re: [linux-audio-dev] One API for everything (first draft)
From: Paul Davis (pbd_AT_Op.Net)
Date: Sat May 19 2001 - 20:28:19 EEST
Even though he doesn't need it, I just want to back Karl up when he writes:
>The semantic of component is limited already - both LADSPA and the
>proposed LAAGA are about the exchange of audio data between components.
Thats precisely it. We need to solve this problem first before
attempting to consider control data and other types of data
streams. It may be that the correct solution to this problem can be
easily extended to cover other non-audio data exchange. It may
not. But the focus needs to be on audio data, since this is the crux
of getting different apps to communicate. Control data is much more
likely to be exchanged between a GUI and an app; MIDI or similar data
can probably be handled by the ALSA virmidi (or pipes, or sockets)
since the latency requirements are rather different.
>I can think of no reason to deal with samples in any format other than
>floats. Previously the use of MMX might have been compelling, but now all
as long as we include the possibilty of "float" having a compile-time
switch to mean either 32 or 64 bit floats, I agree. but only compile
time. there should be no support for any format other than whatever
"float" format was chosen at compile time.
i was reading about the Hammerfall "DSP" yesterday (its not really a
DSP chip; they've just been quite clever at programming the Xilinx
FPGA), and they have some interesting examples of why a 32 bit float
may not be sufficient if we really want "the best" audio fidelity with
24 bit multichannel mixdown.
--p
This archive was generated by hypermail 2b28 : Sat May 19 2001 - 20:52:22 EEST