[LAD] Beta testers required...
tom at trellis.ch
tom at trellis.ch
Thu Jul 24 13:42:00 UTC 2014
If you have synced audio interfaces, this might work too:
It is also a regular jack client, no client/server, one-to-one,
one-to-many (broadcast), for LAN use, UDP (OSC blobs), non-compressed
Has configurable buffer on receiver side, different behaviour models
(there is a 'back' channel), up to 64 voices, different period sizes
BUT.. no resampling, not even when the same SR. So either using a large
buffer on the receiver or use word clock synced audio interfaces.
I've successfully transfered some channels from pc1 to pc2, where both
have a firewire audio interface, one of which is the master, the other is
slaved with S/PDIF. Both pcs run a fully independent JACK (at different
period sizes, same SR). The short S/PDIF cable makes the network case
"local" again, but it works quite reliably (i.e. thight buffer sizes
possible since there is almost no drift). If you give it a try i will be
interested to hear your feedback.
(the audio interfaces drift like this when not synced:
http://lowres.ch/misc/20131228_sisco_spdif_drift.ogv. using robin gareus'
great sisco scope)
On Thu, July 24, 2014 15:17, Jesse Cobra wrote:
> Exactly, the user may already have PTP/802.1as running between systems
> for a common clock... On Jul 24, 2014 3:23 AM, "John Rigg"
> <ladev7 at jrigg.co.uk> wrote:
>> On Thu, Jul 24, 2014 at 08:42:48AM +0000, Fons Adriaensen wrote:
>>> You need resampling even if the sample rates are equal, unless
>>> the interconnected system have a common word clock.
>> Can the resampling be switched off in cases where a common word
>> clock is available?
>> Linux-audio-dev mailing list
>> Linux-audio-dev at lists.linuxaudio.org
> Linux-audio-dev mailing list
> Linux-audio-dev at lists.linuxaudio.org
More information about the Linux-audio-dev