[LAU] open hw soundcard with ext. codec

[Karl Hammar]

Martin Homuth-Rosemann:
> Am Mittwoch, 18. November 2009 schrieb Karl Hammar:
> > Martin Homuth-Rosemann:
...
> I asked you (and the LAU audience) to hear your/their opinion about my (maybe 
> silly) idea. I see this discussion in the early project status mainly as a 
> kind of brainstorming. 

Yes it is, but I has "the other" agenda to design and build a cpu card 
and some slow i/o cards.

I don't consider your questions silly. I try to understand what your
interest is and see what our common goals are.

My roadmap would be something like:

1, design and build the slow i/o card (with lots of i/o)
2, dito cpu card
3, dito a 8 channel 16b/48kHz in/out card
4, dito 24b/96kHz in/out

> To speed up the process of prototyping and to allow many participants I 
> suggested the use of "ready mades" - don't reinvent the wheel! 
> The atmel board ATNGW100 is easy available and not expensive, no time 
> consuming soldering (and hw debuging) needed - this will be a standard 
> platform for colaboration. 

I have that card, we can use it as a common platform.

> The same goes for the ADC, if we use available units with an (open) standard 
> communication protocol like AES-3, ADAT or MADI we can concentrate on the 
> difficult and more exciting part - finding new solutions / algorithms for syncing 
> different sources, internet transfer, ...

Ok, that is good enought, a pity that I don't have any audio equipment 
with thoose interfaces.

...
> > > We need the codec, some kind of amplification, a clean power supply etc.
> > > to get a good S/N ratio - and we need it for a lot of channels.
> > Do you have a spec. which you'd like to discuss ?
> > E.g. how many channels are you regulary using, what s/n ratio is a
> > minimal requirement for you ?
> We (no pluralis maiestatis, I summed my impression from some postings of the 
> LAU audience) need more than two channels, more than 16 bit and more than 
> 100dB S/N at minimal 48 kHz, preferably 96 kHz.

Ok, shall we set the goal to >= 8ch in/out at 24b 96kHz 100dB S/N,
with some inbetween easier targets ?

Is it useful to have 24b/96kHz on output, or would it be sufficient
to have 16bit/48kHz ?

...
> > If this project shall implement any of theese interfaces it might
> > then be the ADAT or MADI, since I see no reason to implement the
> > smaller interfaces.

> But AES-3 (or AES-42 for digital microphone) is a standard for digital audio 
> connection.

But I thought that the goal was to have more than mere 2-4 channels ?

> > But if we successfully implement adat or madi, we are still missing
> > the adat/madi part on the pc. So we still have a problem...
> No, our "LAU-interface" is this part on the pc.

I don't understand you. Can you give an example?

***

I see two different things here:

A, a hardware project

B, finding new solutions/algorithms for syncing different sources...

***

My main interest is project A.

For project B, can't we simply use ordinary pc's, one master and
a few slaves, or am I missing something ?

Initially I had the idéa of a very small card with a cpu and one
adc/preamp, a mic-card, but that idéa didn't receive much interest.
A few of thoose hypothetical mic-card plus a master pc could be a
platform for testing syncing problems/solutions and distributed
recording. I'd consider that an interesting problem.

Regards,
/Karl

-----------------------------------------------------------------------
Karl Hammar                    Aspö Data               karl at aspodata.se
Lilla Aspö 148                                                 Networks
S-742 94 Östhammar          +46  173 140 57                   Computers
Sweden                     +46  70 511 97 84                 Consulting
-----------------------------------------------------------------------