Subject: Re: [linux-audio-dev] ALSA vs OSS/free
From: David Olofson (david_AT_gardena.net)
Date: Sun Mar 10 2002 - 20:37:26 EET
On Sunday 10 March 2002 18.34, Paul Davis wrote:
[...]
> half-baked solution to a genuine problem. The design is the same as
> the win32 kernel mixer, which even MS now admits was a bad idea.
You mean the latency is not a *feature*!? Wow... *wonders if MS is
changing, or something*
[...]
> >Just set it to something like 8 and see value returned in channel
> > count.
>
> And if the correct answer is 26? or 52, if I've merged two
> Hammerfalls together? I'm not even sure OSS has even bits
> available for really high channel counts.
...and it has only 16 bits for fragment size...
[...]
> ALSA0.5 is obsolete and should not, in my opinion be supported by
> anyone or anything at this time. i have never heard of comedi.
"Comedi" is a data acquisition driver framework (+ drivers) for
RTL/RTAI (works with both, IIRC). AFAIK, it cannot be used without a
hard real time extension, although I could be wrong about that. (It's
not a major problem writing drivers that compile with or without RTL
or RTAI, as long as the design is reasonably sound.)
Either way, it's not meant for normal audio cards - but there are
many parallels. For example, most DAQ cards have *lots* of channels -
and some of them operate at >Msamples/s rates, capable of pushing any
system to it's limits, even when using busmaster DMA. Not all that
different from high end multichannel audio interfaces...
[...]
> >I don't think RT-Linux and RT-AI as different operating systems.
> > Only as realtime extensions to Linux kernel that could be
> > included in standard kernel.
>
> They need new drivers. In my mind, thats as good/bad as a new OS.
Right. I've written code that bridge the Linux/RTL gap, and it's not
pretty. The major issue is that Linux synchronization constructs are
worth nothing in RTL code, so a real time driver needs to take
special measures to interface with the kernel.
This can be solved, of course, but that basically means that you need
two layer drivers; one RT part and one Linux part, with an interface
between them based on RTL/RTAI sync constructs.
(Then again, even that environment is helluvalot nicer than what
Windows NT driver coders have to deal with, from what I've heard...
;-)
//David
.- M A I A -------------------------------------------------.
| Multimedia Application Integration Architecture |
| A Free/Open Source Plugin API for Professional Multimedia |
`----------------------> http://www.linuxaudiodev.com/maia -'
.- David Olofson -------------------------------------------.
| Audio Hacker - Open Source Advocate - Singer - Songwriter |
`-------------------------------------> http://olofson.net -'
This archive was generated by hypermail 2b28 : Sun Mar 10 2002 - 20:27:51 EET