Hi Patrick,
That's great! Does it still rely on system clocks being in sync via
ntp or similar? Or does it do that itself? It seems to do that
looking at the code, but I can't really read sclang.
I guess bad wifi will always be a problem for time sync, impressive
that you got it robust on any kind of wifi really.
It would be good to either make it peer to peer, where the master
clock is negotiated, or have a daemon that is implemented in C with
minimal dependencies. My stuff is all dependent on supercollider at
the moment which makes it a bit trickier to package/get working.
alex
On 9 February 2011 15:45, Patrick Borgeat <patrick_at_borgeat.de> wrote:
> Hi there,
>
> No proposal for a standard but I thought I let you all know, if somebody
> wants to try it out:
>
> We coded our own sync/communication system for SuperCollider (see
> MandelClock). It's good and robust on a decent WiFi (or LAN) but sometimes
> wents a bit crazy on a bad WiFi. We use it a lot at the moment.
>
> One thing a band has to do though is to adjust Server latencies, (s.latency)
> to compensate for different audio drivers and stuff. Just fire up m.metro
> and adjust the latencies till every player sounds in sync.
>
> cheers,
> Patrick
>
> Am 09.02.2011 um 16:29 schrieb alex:
>
>> It would still be nice getting a standard + simple to use way of
>> syncing across live coding environments. Maybe Andrew's new extempore
>> project could give a needed push in this direction?
>
>
--
http://yaxu.org/
Received on Wed Feb 09 2011 - 15:57:02 GMT