Andreas do let me know if you ever get round to writing that, sounds
interesting. I ended using the NetUtil library (
http://www.sciss.de/
netutil/), for which this hasn't been an issue at all. I tried to
debug the JavaOSC library, but couldn't make any headway with it. The
issue only seemed to happen in high-traffic situations which makes
tracking down the exact point tricky...
On Aug 6, 10:55 am, Andreas Reuterberg <
andreas.reuterb...@gmail.com>
wrote:
> .. oh, and perhaps a OSC library which can handle both UDP and TCP.
>
> On Fri, Aug 6, 2010 at 10:53 AM, Andreas Reuterberg
>
>
>
> <
andreas.reuterb...@gmail.com> wrote:
> > I've used the same OSC library for FingerPlay MIDI without any
> > problems. You can have a look at my source code if you want.
>
> >
http://code.google.com/p/fingerplaymidi
>
> > I wish I had the time to write a "memory optimized" OSC library for
> > Android, one which doesn't do a 'new' for each packet sent :).
>
> > /Andreas
>
> > On Fri, Aug 6, 2010 at 8:13 AM, Dan Stowell <
danstow...@gmail.com> wrote:
> >> OSC is not made on top of UDP, it's agnostic of how it's delivered. In
> >> supercollider for example you can choose TCP or UDP. The issue kind of
> >> sounds a bit like an internal bug in javaOSC to me (race condition?),
> >> but as José suggests it could be UDP's fault - so if you're able to
> >> switch to TCP instead, that might fix it.
>
> >> Dan
>
> >> 2010/8/6 Jip <
jipa...@gmail.com>: