29 Oct
2011
29 Oct
'11
9:43 a.m.
> A suggestion: start with providing just digital
interfaces: ADAT, and
> MADI if you're ambitious. Both are quite straightforward. ...
> I wouldn't forget poor old AES (including SPDIF.)
How about a standard adc stage with FPGA processing/routing and I2S bus. For ADAT there
are already the OPTOGEN and OPTOREC modules (dont have the links right now). Then ADAT and
spdiff/aes-ebu as standard interfaces. An optional XMOS low cost board for avb, which is
capable of I2S. And a FPGA extension board for MADI and ieee1394 also connected via I2S.
I kinda like the idea... could be highly flexible...
Second step: same device as DAC...
Sorry fucked up the subject...
Ck
29 Oct
29 Oct
3:58 p.m.
New subject: Open Sound Interface project beginnings
On Sat, Oct 29, 2011 at 2:20 AM, Christoph Kuhr <christoph.kuhr(a)web.de> wrote:
I'm still not clear on it. I was having a little trouble following
the ideas from Dale and Fons, too. How would digital audio interfaces
be different from adc/dac?
The data transfer to/from the PC is packetized, no matter what kind of
interface you use to transfer data. It's an asyncrhonous process that
depends on when data is sent to or ready to be read from the device.
My idea is to control adc/dac directly from an FPGA (lots of them,
potentially). I'm familiar with I2C and SPI interfaces on adc/dac,
and there's a few differences between devices (order of bits from MSB
to LSB or LSB to MSB for instance). They are all basically the same,
but the handling of the same data is a bit different.
I have a few questions
Is digital audio syncronized or packetized? What's the clock rate?
What's the data transfer bandwidth? Is there an IC you can use just
like adc/dac?
Chuck
> A
suggestion: start with providing just digital interfaces: ADAT, and
> MADI if you're ambitious. Both are quite straightforward. ...
> I wouldn't forget poor old AES (including SPDIF.)
How about a standard adc stage with FPGA processing/routing and I2S bus. For
ADAT there are already the OPTOGEN and OPTOREC modules (dont have the links
right now). Then ADAT and spdiff/aes-ebu as standard interfaces. An optional
XMOS low cost board for avb, which is capable of I2S. And a FPGA extension
board for MADI and ieee1394 also connected via I2S.
I kinda like the idea... could be highly flexible...
Second step: same device as DAC...
Sorry fucked up the subject...
Ck
5:51 p.m.
New subject: Open Sound Interface project beginnings
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
My idea is to control adc/dac directly from an FPGA
(lots of them,
potentially). I'm familiar with I2C and SPI interfaces on adc/dac,
and there's a few differences between devices (order of bits from
MSB to LSB or LSB to MSB for instance).
yes, i thought so too...
something like this:
______________
| |<-> *2 <->adat
adc *1 ->| |<-> *2 <->spdif/aes-ebu
|fpga ->i2s |<->xmos avb
dac *1 <-| *3 |<->some phy <->madi
|_____________|<->some phy <->ieee1394
*1 you need external adc and dac, because lowcost developement onboard
adc/dac have only 10 or 12bit resolution...
*2 adat to I2S boards (properiterian codec)
http://electronics.dantimax.dk/Kits/Digital_audio/11329401182.html
spdif/aes-ebu to I2S boards
http://electronics.dantimax.dk/Kits/Digital_audio/11319920832.html
http://electronics.dantimax.dk/Kits/Digital_audio/12906353062.html
*3 MSB/LSB switching can easyly be done with fpga
Ck
- --
PGP Key ID 0x528422c1;
PGP Public Key / PGP Key verification:
pgp.mit.edu
pgp.zdv.uni-mainz.de
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iQEcBAEBAgAGBQJOrCESAAoJEBra75lShCLBqaUH/RHn4XO9iP+xUaaFYknXBPIt
NNWq6VFCm39Wk/l+0lWYby7H3X/T5fiwPHG0/rKgekgklCauhLAmy6j3pjocJ3lL
9K6Zrnq+poGEPqRPu1qdU2m9SBB9+Yw/bZb0BXNI8ZABvTuu25Y2sPldis/E/bFn
9rY2JSnS3dNiZfq3bEO9wyfe+5TPJsACaYOJpXlSJCiC0q6ijQG3bZk7wTWWArqx
8amMMP+c6NbZHra4vMVawbUbzBP6k1HTwyEuDnneGzl6Uf4FNjWA02ygw7ei1Tnm
VAbXG6A/Cr/N5gDv1WMVyVI231EzJuY56FdLzenXXsWVJfDhwmujD3J5e5h4GOc=
=uHK2
-----END PGP SIGNATURE-----
10:50 p.m.
New subject: Open Sound Interface project beginnings
On Sat, Oct 29, 2011 at 08:58:46AM -0500, Charles Henry wrote:
I have a few questions
Is digital audio syncronized or packetized? What's the clock rate?
What's the data transfer bandwidth?
There's no single answer to that. ADAT (AFAIK) is synchronised,
but MADI certainly isn't as it can allow variable sample rate.
For MADI the logical bit rate is 100 MHz, or 125 MHz physical
due to 4B5B encoding (which allows recovery of the clock at the
receiver), and this is independent of the actual sample rate.
Any remaining capacity is filled by sync words. This is used
(at 44.1 or 48 kHz) either for 56 channels with 'varispeed'
up to +12.5%, or for 64 channels without the varispeed (the
latter is how I guess it is used in most installations). For
96 kHz the number of available channels is halved. Samples
are 32 bit, with 24 bits used for the actual audio data and
8 for various signalling bits.
Is there an IC you can use just like adc/dac?
No idea. But the format isn't complicated at all.
Ciao,
--
FA
4777
days inactive
4777
days old
linux-audio-user@lists.linuxaudio.org
3 comments
3 participants
participants (3)
-
Charles Henry -
Christoph Kuhr -
Fons Adriaensen