26 Oct
2011
26 Oct
'11
2:56 p.m.
Hi!
great idea!!!!!!!!
im desperatly looking for AVB (IEEE 802.1BA) adc/dac s. this could be
THE solution...
1. Modules for dac and adc with on-board identifiers
(mixed-signal design)
2. A FPGA-based programmable system board with connectors for
respective modules (high-frequency circuit design)
3. FPGA code for buffers, clocks, and device discovery (VHDL/Verilog
programming)
Did you have a look at opencores.org for IPs?
I2C and I2S cores are available at opencores.org
4. Data transport modules (FPGA code plus hardware
design), could be
USB, FW, ETH, PCI, wireless, etc...
Perhapes two transport modules?
like FW and ETH -> AudioVideoBridging? (they share the same codec
implementation: iec61883-x)
if you like to use it with a laptop, use FW. if you have a little
projectstudio with seperated recording/control room situation it would
integrate in an AVB environment.
The DAW would need to have an AVB ready NIC, means PTP HW timestamping
between PHY and MAC.
At the moment im working on a linux implementaion for AVB, perhaps this
is what i could contribute?
my implementation will be jack a application...
There is an AVB properitarien IP for FPGA, apparently 25000€.
=> there would be the need for opensource implementation.
so i would like to contribute my AVB stuff if/when i got it working ->
estimated deadline july 2012...
after that i would like to have a look at an open AVB IP for FPGA.
bye
Ck
26 Oct
26 Oct
5:38 p.m.
New subject: Open Sound Interface project beginnings
On Wed, Oct 26, 2011 at 7:56 AM, Christoph Kuhr <christoph.kuhr(a)web.de> wrote:
That seems pretty interesting, but it could be more streamlined. RISC
is (I think) overkill, but it wouldn't hurt to study it :)
http://opencores.org/project,i2c
Gigabit ethernet has become ubiquitous on new hardware at about the
same time that firewire has become rare... I think mostly we could use
just one interface at a time---but reuse a lot of the code and
hardware.
...
Hi!
great idea!!!!!!!!
im desperatly looking for AVB (IEEE 802.1BA) adc/dac s. this could be THE
solution...
Hi Christoph
I'm unfamiliar with AV bridging--what do you use it for?
1. Modules for dac and adc with on-board
identifiers (mixed-signal
design)
2. A FPGA-based programmable system board with connectors for
respective modules (high-frequency circuit design)
3. FPGA code for buffers, clocks, and device discovery (VHDL/Verilog
programming)
Did you have a look at opencores.org for IPs?
I2C and I2S cores are available at opencores.org 4. Data transport modules (FPGA code plus
hardware design), could be
USB, FW, ETH, PCI, wireless, etc...
Perhapes two transport modules?
like FW and ETH -> AudioVideoBridging? (they share the same codec
implementation: iec61883-x)
if you like to use it with a laptop, use FW. if you have a little
projectstudio with seperated recording/control room situation it would
integrate in an AVB environment.
The DAW would need to have an AVB ready NIC, means PTP HW timestamping
between PHY and MAC.
At the moment im working on a linux implementaion for AVB, perhaps this is
what i could contribute?
my implementation will be jack a application
There is an AVB properitarien IP for FPGA, apparently 25000€.
=> there would be the need for opensource implementation.
so i would like to contribute my AVB stuff if/when i got it working ->
estimated deadline july 2012...
after that i would like to have a look at an open AVB IP for FPGA.
Isn't it crazy how expensive some of the FPGA boards and software IP
becomes? It's like it's own little anti-competitive niche market
*cough* DoD *cough*
While the FPGA's themselves aren't prohibitively expensive, the rapid
development boards+software are.
6:01 p.m.
New subject: Open Sound Interface project beginnings
I'm unfamiliar with AV bridging--what do you use
it for?
AVB is a !!! layer2 !!! ethernet standard for audio and video transport,
with precise timing synchronization (house clock like with grandmaster
clock selection), resource reservation and quality of service.
worstcase latency over 7hops: 2ms
needs fast eth at least.
the pro audio industry is presenting first products soon, meyersound
already has it in use in its dmitri.
harman pro (studer, soundcraft, bss, jbl,...), yamaha will support it,
biamp, focusrite, audinate, and many many more...
this compatibility would make such an open interface much more
interesting, i think.
While the FPGA's themselves aren't
prohibitively expensive, the rapid
development boards+software are.
i have a low budget board from xilinx (200$), but
will soon switch to
altera, because the are cheaper boards available (60€)...
bye
Ck
6:23 p.m.
New subject: Open Sound Interface project beginnings
On Wed, Oct 26, 2011 at 11:01 AM, Christoph Kuhr <christoph.kuhr(a)web.de> wrote:
OK, that's a very good standard. Then, it's also addressable, so you
could have multiple audio devices on the same network. It should also
be good for designing large PA systems (over whole buildings, concert
halls, arenas, etc...) using standard networking.
I have a Basys2 Xilinix board to get started on. It's decently priced
and has an I2C DAC module (20$ extra) for me to play with, plus some
breadboards for trying out other IC's. At my university, I can take
an FPGA course next spring also. Seems like it's still a long road
ahead before having anything interesting programmed.
I'd also be interested in other manufacturers--the big issue for me is
figuring out which programming interface works best (on Linux
preferrably!!!). The Xilinx ISE 13.2 needed a little debugging of
shell scripts, so far.
I'm unfamiliar with AV bridging--what do you
use it for?
AVB is a !!! layer2 !!! ethernet standard for audio and video transport,
with precise timing synchronization (house clock like with grandmaster clock
selection), resource reservation and quality of service.
worstcase latency over 7hops: 2ms
needs fast eth at least. the pro audio industry is presenting first products
soon, meyersound already
has it in use in its dmitri.
harman pro (studer, soundcraft, bss, jbl,...), yamaha will support it,
biamp, focusrite, audinate, and many many more...
this compatibility would make such an open interface much more interesting,
i think.
Where does the implementation belong? It ought to factor into
designing kernel modules + hardware (eth) + fpga code.
While the
FPGA's themselves aren't prohibitively expensive, the rapid
development boards+software are.
i have a low budget board from xilinx (200$), but will soon switch to
altera, because the are cheaper boards available (60€)...
bye
Ck 
7:28 p.m.
New subject: Open Sound Interface project beginnings
On Wed, 26 Oct 2011 11:23:53 -0500
Charles Henry <czhenry(a)gmail.com> wrote:
On Wed, Oct 26, 2011 at 11:01 AM, Christoph Kuhr
<christoph.kuhr(a)web.de> wrote:
OK, that's a very good standard. Then, it's also addressable, so you
could have multiple audio devices on the same network. It should also
be good for designing large PA systems (over whole buildings, concert
halls, arenas, etc...) using standard networking.
I have a Basys2 Xilinix board to get started on. It's decently priced
and has an I2C DAC module (20$ extra) for me to play with, plus some
breadboards for trying out other IC's. At my university, I can take
an FPGA course next spring also. Seems like it's still a long road
ahead before having anything interesting programmed.
I'd also be interested in other manufacturers--the big issue for me is
figuring out which programming interface works best (on Linux
preferrably!!!). The Xilinx ISE 13.2 needed a little debugging of
shell scripts, so far.
Wasn't this very much like what was discussed about a year ago?
There is this:
http://www.xmos.com/avbl2
The code for this reference design is royalty free, and lots of detail is
available. With all the documentation and code available, surely it would be
possible to make a linux equivalent to an endpoint.
These are BGA devices so not suitable for D.I.Y. builds, but I wonder how much a
small production run would be for a board with just the power, Ethernet and I2S
headers.
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
I'm unfamiliar with AV bridging--what do you
use it for?
AVB is a !!! layer2 !!! ethernet standard for audio and video transport,
with precise timing synchronization (house clock like with grandmaster clock
selection), resource reservation and quality of service.
worstcase latency over 7hops: 2ms
needs fast eth at least. the pro audio industry is presenting first
products soon, meyersound already
has it in use in its dmitri.
harman pro (studer, soundcraft, bss, jbl,...), yamaha will support it,
biamp, focusrite, audinate, and many many more...
this compatibility would make such an open interface much more interesting,
i think.
Where does the implementation belong? It ought to factor into
designing kernel modules + hardware (eth) + fpga code.
While the
FPGA's themselves aren't prohibitively expensive, the rapid
development boards+software are.
i have a low budget board from xilinx (200$), but will soon switch to
altera, because the are cheaper boards available (60€)...
bye
Ck 
7:58 p.m.
New subject: Open Sound Interface project beginnings
Regarding the USB target, I also know of this project
http://www.linux-usb-daq.co.uk/
which may be helpful.
It's a USB data acquisition board which works with the COMEDI drivers.
Everything is open and you may find something helpful there.
On 10/26/11, Folderol <folderol(a)ukfsn.org> wrote:
On Wed, 26 Oct 2011 11:23:53 -0500
Charles Henry <czhenry(a)gmail.com> wrote:
On Wed, Oct 26, 2011 at 11:01 AM, Christoph Kuhr
<christoph.kuhr(a)web.de>
wrote:
OK, that's a very good standard. Then, it's also addressable, so you
could have multiple audio devices on the same network. It should also
be good for designing large PA systems (over whole buildings, concert
halls, arenas, etc...) using standard networking.
I have a Basys2 Xilinix board to get started on. It's decently priced
and has an I2C DAC module (20$ extra) for me to play with, plus some
breadboards for trying out other IC's. At my university, I can take
an FPGA course next spring also. Seems like it's still a long road
ahead before having anything interesting programmed.
I'd also be interested in other manufacturers--the big issue for me is
figuring out which programming interface works best (on Linux
preferrably!!!). The Xilinx ISE 13.2 needed a little debugging of
shell scripts, so far.
Wasn't this very much like what was discussed about a year ago?
There is this:
http://www.xmos.com/avbl2
The code for this reference design is royalty free, and lots of detail is
available. With all the documentation and code available, surely it would be
possible to make a linux equivalent to an endpoint.
These are BGA devices so not suitable for D.I.Y. builds, but I wonder how
much a
small production run would be for a board with just the power, Ethernet and
I2S
headers.
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
I'm unfamiliar with AV bridging--what do you
use it for?
AVB is a !!! layer2 !!! ethernet standard for audio and video transport,
with precise timing synchronization (house clock like with grandmaster
clock
selection), resource reservation and quality of service.
worstcase latency over 7hops: 2ms
needs fast eth at least. the pro audio industry is presenting first
products soon, meyersound
already
has it in use in its dmitri.
harman pro (studer, soundcraft, bss, jbl,...), yamaha will support it,
biamp, focusrite, audinate, and many many more...
this compatibility would make such an open interface much more
interesting,
i think.
Where does the implementation belong? It ought to factor into
designing kernel modules + hardware (eth) + fpga code.
While the
FPGA's themselves aren't prohibitively expensive, the rapid
development boards+software are.
i have a low budget board from xilinx (200$), but will soon switch to
altera, because the are cheaper boards available (60€)...
bye
Ck
10:43 p.m.
New subject: Open Sound Interface project beginnings
Where does the implementation belong? It ought to
factor into
designing kernel modules + hardware (eth) + fpga code.
Im at a linux kernel developement workshop. Discussed parts of my design.
I want to do some kernel modules and userspace as well. Also a library using the 61883
codec.
Perhapes we should take this discussion to the lad-list...
I'd also be interested in other manufacturers--the
big issue for me is
figuring out which programming interface works best (on Linux
preferrably!!!). The Xilinx ISE 13.2 needed a little debugging of
shell scripts, so far.
Since i can get support for my university, i will use altera. I chose xilinx before,
because the have the avb ip.........
Bye
Ck
11:11 p.m.
New subject: Open Sound Interface project beginnings
Perhaps we should take this discussion to the
lad-list...
The discussion is interesting, if you don't take it uniquely to LAD then
the group should minimally be copied in. What you are discussing here
really is more D than U.
Kind regards, cnik
"we have to make sure the old choice [Windows] doesn't disappear”.
Jim Wong, president of IT products, Acer
From: christoph.kuhr(a)web.de
Date: Wed, 26 Oct 2011 22:42:37 +0200
To: czhenry(a)gmail.com; linux-audio-user(a)lists.linuxaudio.org
Subject: Re: [LAU] Open Sound Interface project beginnings
Where does the implementation belong? It ought to
factor into
designing kernel modules + hardware (eth) + fpga code.
Im at a linux kernel developement workshop. Discussed parts of my design.
I want to do some kernel modules and userspace as well. Also a library using the 61883
codec.
Perhapes we should take this discussion to the lad-list...
I'd also be interested in other
manufacturers--the big issue for me is
figuring out which programming interface works best (on Linux
preferrably!!!). The Xilinx ISE 13.2 needed a little debugging of
shell scripts, so far.
Since i can get support for my university, i will use altera. I chose xilinx before,
because the have the avb ip.........
Bye
Ck
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
27 Oct
27 Oct
11:17 p.m.
New subject: Open Sound Interface project beginnings
On Wed, Oct 26, 2011 at 4:11 PM, Nick Copeland
<nickycopeland(a)hotmail.com> wrote:
I agree. I appreciate the feedback from everyone in the discussion
(and anyone else who'd like to discuss further). This is sort of a
brainstorming session--feel free to keep the thread alive until you
get bored :)
Although I've been playing with the idea and studying a bit for the
past year, I haven't wanted to take this idea to anyone without having
something to back it up. Regardless, it's about time to start mapping
things out and ordering what to study and what to implement.
Perhaps we
should take this discussion to the lad-list...
The discussion is interesting, if you don't take it uniquely to LAD then
the group should minimally be copied in. What you are discussing here
really is more D than U.
Kind regards, cnik 
11:45 p.m.
New subject: Open Sound Interface project beginnings
On Thu, Oct 27, 2011 at 04:17:50PM -0500, Charles Henry wrote:
I agree. I appreciate the feedback from everyone in
the discussion
(and anyone else who'd like to discuss further). This is sort of a
brainstorming session--feel free to keep the thread alive until you
get bored :)
Although I've been playing with the idea and studying a bit for the
past year, I haven't wanted to take this idea to anyone without having
something to back it up. Regardless, it's about time to start mapping
things out and ordering what to study and what to implement.
A suggestion: start with providing just digital interfaces: ADAT, and
MADI if you're ambitious. Both are quite straightforward. I'd expect
them anyway, and it avoids all the troubles of mixed analog/digital
PCB design which can be very hard, in particular for audio.
Ciao,
--
FA
28 Oct
28 Oct
12:48 a.m.
New subject: Open Sound Interface project beginnings
Date: Thu, 27 Oct 2011 21:45:31 +0000
From: fons(a)linuxaudio.org
To: linux-audio-user(a)lists.linuxaudio.org
Subject: Re: [LAU] Open Sound Interface project beginnings
On Thu, Oct 27, 2011 at 04:17:50PM -0500, Charles Henry wrote:
I wouldn't forget poor old AES (including SPDIF.) Some of us may not want to work with
a minimum of eight channels.
I agree. I appreciate the feedback from everyone
in the discussion
(and anyone else who'd like to discuss further). This is sort of a
brainstorming session--feel free to keep the thread alive until you
get bored :)
Although I've been playing with the idea and studying a bit for the
past year, I haven't wanted to take this idea to anyone without having
something to back it up. Regardless, it's about time to start mapping
things out and ordering what to study and what to implement.
A suggestion: start with providing just digital interfaces: ADAT, and
MADI if you're ambitious. Both are quite straightforward. I'd expect
them anyway, and it avoids all the troubles of mixed analog/digital
PCB design which can be very hard, in particular for audio.
Ciao,
--
FA
10:48 a.m.
New subject: Open Sound Interface project beginnings
On Thu, Oct 27, 2011 at 10:48:37PM +0000, Dale Powell wrote:
I wouldn't forget poor old AES (including SPDIF.)
Some of us may not want to work with a minimum of eight channels.
Agreed. Cards with AES interfaces are hard to find and
are usually expensive.
Ciao,
--
FA
4575
days inactive
4577
days old
linux-audio-user@lists.linuxaudio.org
11 comments
7 participants
participants (7)
-
Charles Henry -
Christoph Kuhr -
Dale Powell -
Egor Sanin -
Folderol -
Fons Adriaensen -
Nick Copeland