Subject: Re: [linux-audio-user] open source firewire
From: Mark Knecht (markknecht_AT_gmail.com)
Date: Mon Nov 29 2004 - 00:26:16 EET
On Sun, 28 Nov 2004 16:12:44 -0600, Ryan Gallagher
<ruinaudio_AT_comcast.net> wrote:
> On Sun, 2004-11-28 at 16:20 -0500, Dave Robillard wrote:
>
>
> > On Sun, 2004-28-11 at 10:55 -0600, Ryan Gallagher wrote:
> > > Anyway, to get to the point... isn't it true that firewire by virtue of
> > > it's similar-to-USB-nature suffers from the same limitations as USB?
> > > Iirc I was told that whatever limitations causing the (IMO) extreme
> > > suckage of USB audio on linux will be present in a firewire interface.
> >
> > Firewire isn't very similar to USB /at all/, other than the fact that
> > they're external busses that can be used to connect things to a
> > computer.
> >
> > Look at some information about the architechture of each system, they're
> > drastically different (Firewire being vastly superior in every single
> > way, for the record)
> >
> > For one thing, Firewire doesn't require a host computer at all, whereas
> > USB of course does. Firewire devices will figure out the layout of the
> > firewire 'network' and establish appropriate master/slave relationships
> > all by themselves, no computer needed. That should give an idea how
> > similar they are.
>
> Of course there are major differences. But there *are* similarities
> beyond those listed here.
>
> I'm no expert but this cryptic note from a jack developer hints at the
> USB issues... I recall someone saying the same is true for FireWire.
>
> "...problem is due to JACK demanding power of two buffer sizes while USB
> devices typically demand an even number of milliseconds. This causes the
> sound to break up when running at even moderately low latencies." - joq
>
> So apparently it's about how data is transmitted, milliseconds vs.
> size.
>
> This is all conjecture on my part, I'm not a programmer type.
>
Ryan,
I'm not a programmer either so you and I are in the same boat.
However, and with the greatest respect for the developers of Alsa,
Jack and all that goes with it, just using logic it seems to me that
this statement cannot be an all encompassing answer for the problems
you are seeing. If it was irrefutably true then USB audio wouldn't
work under windows or Mac OS's either, would it?
It seem to me, and just speaking from a logical perspective and not
a techncial one, that if USB audio cannot be made to work well under
Linux when it does (apparent;y) work well under Windows and Mac OS's
that there must be a problem with either Alsa, Jack or Linux itself. I
cannot say which one, but I think my statement is true from a lagical
perspective.
- Mark
This archive was generated by hypermail 2b28 : Mon Nov 29 2004 - 00:35:25 EET