[linux-audio-dev] Open firewire audio interface:
A back-of-an-envelope prototype plan
sjenkins at blueyonder.co.uk
Sat Nov 27 00:20:38 EST 2004
Bob Knight wrote:
> I want an open ethernet audio device.
> I do not want or need to deal with pci, firewire, usb.
Fair enough, and an interesting project, though personally I'm
more interested in doing a firewire device.
> Use one of many controllers out there with a good embedded
> linux port and build in ethernet.
Cirrus do some ARM 9 based SoCs with built-in Ethernet MAC.
They run linux given enough flash & RAM, but eCos would
probably be a better fit RTOS for this application IMO. They've
got on-chip AC'97 interface too if someone wants to do a 2-in
2-out design, or to start with one whilst prototyping.
> Use a TDM or GPIO type
> interface for the analog side of the world.
I was thinking of clocking the data into and out of some D/As
and A/Ds, ie solving the problem rather than moving it to the
other end of a cable and putting a different connector on it.
In case it wasn't clear the Converter Interface Engine on my
block diagram is intended to clock data into and out of audio
converters. The Dig Audio Interface header takes these signals
off the prototyping board to make development cheaper and
> TDMoE would work just fine for audio.
> One thing that linux does well these days is ethernet.
> I need a minimum of 32 channels for live recording.
> The idea would be to just suck in the ethernet frames and write them
> to disk. Then mix them down back in the studio after the show.
I see now why you ask for something "GPIO like": You want to be able
to put, say, 4 * 8-channel input modules on it to get exactly the I/O - or
just the "I" in this case - that you require.
> I am a noop (programmer) with a EE back ground and would glad to
> help anyone build something like this.
OK. I'm employee & minority shareholder in a small company doing
industrial embedded control h/w and s/w development. I've also worked
on multi-channel audio I/O hardware in the past. So I could let myself
into work on Sunday and start laying out a board if I wanted to, then I
could get it built, then I could get it running. If it was worth the time
and money. And if I could spare the time and money.
I'm not claiming to know it all though and some help would be
required. My FPGA experience is limited and out of date, I've
never worked with firewire, I don't do analog circuit design (I can
copy them though, and I can do mixed-signal PCB layout) and I've
never worked on linux drivers. Part of my motivation here is to
get some development experience in some of these areas.
I've got relatively little time or money to spare which is why I'm
looking for ways to make the initial development phase as quick
and as cheap as possible without compromising a finished design.
Some cheap early prototypes would also be essential for bringing
other people into the s/w and ip development. Hopefully a running
prototype would then attract the resources to be developed further.
If not then the prototypes would still be pretty neat little FPGA dev
& prototyping boards in their own right. Nice to own and possibly
even sellable for that purpose.
But like I said, I'm more interested in doing it over firewire.
More information about the linux-audio-dev