<quote who="eskogen@usfamily.net"> > will OSC ever support tcp? let's say i wanted to send data to my friend > so-and-so across the internet to such-and-such place. i'd like it to be > OSC and i'd like to use Pd. is there a way to do this? seems like if we > can have udp, we should be able to have tcp with a minumum of effort,
OSC is itself a protocol which is completely independent of the choice of transport method. UDP is most common because it gives the lowest latency for local area networks. Using TCP, or even Jabber, IRC, http, email or shortwave radio is up to the implementor. :)
Another way to get UDP over TCP is via tunneling, e.g. VPN.
however, i had another related question. how can one assure consistant
real-time performance beyond net-lag? are timetag messages reliable for this? i'm just getting interested in OSC so any help would be welcome.
Yes. That is exactly the the purpose of the timetag. If a message is recieved with a timetag which is expired, it gets dropped. Timetags marked in the future are to be queued and held until the specified time. This will enable jitter correction with a hard latency threshold. Note that OSC is intended to be a completely stateless protocol, so dropping packets if latency bounds are not met is considered the right thing to do.
However, I have not yet seen any implementation of OSC that actually uses the timetag feature. Once again, it would not be hard, its just that no one has seen the need to do it... yet.
Andy.
Andrew W. Schmeder andy_at_a2hd_dot_com http://www.a2hd.com
On Sunday 16 March 2003 03:00, Andy Schmeder wrote:
OSC is itself a protocol which is completely independent of the choice of transport method.
OSC seems really interesting, but is it a smart thing not to specify the transport protocol? So far I've seen only one hardware synth (Reaktor) that supports it. Is any commercial manufacturer really going to replace MIDI with this if the transport method isn't specified? I suppose that Ethernet over UDP would be a logical choice, but it would still be somewhat of a gamble on the part of the first manufacturer to implement it, hoping that the others would follow suit.
Or am I missing something here?
Larry Troxler
I guess soft synth come first, then OSC over UDP/IP/Ethernet physical controls (maybe with MIDI too), then hard synth.
This is partially happening/ed.
Maurizio Umberto Puxeddu.
On Sun, 2003-03-16 at 15:13, Larry Troxler wrote:
On Sunday 16 March 2003 03:00, Andy Schmeder wrote:
OSC is itself a protocol which is completely independent of the choice of transport method.
OSC seems really interesting, but is it a smart thing not to specify the transport protocol? So far I've seen only one hardware synth (Reaktor) that supports it. Is any commercial manufacturer really going to replace MIDI with this if the transport method isn't specified? I suppose that Ethernet over UDP would be a logical choice, but it would still be somewhat of a gamble on the part of the first manufacturer to implement it, hoping that the others would follow suit.
Or am I missing something here?
Larry Troxler
PD-list mailing list PD-list@iem.kug.ac.at http://iem.kug.ac.at/cgi-bin/mailman/listinfo/pd-list