[Clam-devel] Dome Yo Frankie

Natanael Olaiz nolaiz at gmail.com
Thu Sep 17 00:23:42 PDT 2009


Hi Pete,

El 09/16/2009 08:07 PM, Pete Carss escribió:
> I tracked down the silence...
>
> In BGEOSCSender.py, the port is being incremented, and was ending up 
> around 7011, 7012 - I've hacked it variously to use port 7000....
>
> I'm a little confused though, is this written for the suggested use of 
> OSC and CLAM? If so, what's the logical definition of a port, ie one 
> port per character, or one port per type of activiy...? With the 
> appropriate listener in CLAM.

At the time that the scripts were created, the OSC CLAM receivers (and 
the used liblo library) doesn't allow multicast servers (i.e.: allow to 
exist several OSC receiver servers in different threads in the same 
port) ... That means that if we want to use several sound sources, they 
should be in the same CLAM network, that could be very big... So: as we 
want to deal different sources in different networks, we used different 
ports for the sources.... And as you need the listener 
position/orientation too, the listener sends that data to all the ports.

Now you can use the same port in different networks if you install the 
latest liblo library and enable the multicast option on the receivers... 
Or as I imagine you did, just use one big network.

>
> or
>
> Am I missing something about the CLAM objects? When you specify say 
> port 7000, are you just specifying the beginning of a range of ports 
> for that object to listen on...?
No, the receivers just use the port that you define.

>
> Is this documented somewhere...?
>

Sorry, not ATM. :-/

But feel free to ask again if you have more questions.


Best regards,
Natanael.

PS: I'm testing the new Blender 2.5, so I expect to actualize the 
scripts to be compatible with it / Python 3.1 soon.

> Pete
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Clam-devel mailing list
> Clam-devel at llistes.projectes.lafarga.org
> https://llistes.projectes.lafarga.org/cgi-bin/mailman/listinfo/clam-devel
>   





More information about the clam-devel mailing list