On Tue, Jan 15, 2008 at 03:38:02PM +0000, Colin Guthrie wrote: > Ulf Behrens wrote: > > PA source -> socket -> netcat -> socket -> PA sink via the simple > > protocol > > or > > PA source -> gstreamer -> PA sink > > > > or any other idea? > > Well you could definately do it via gstreamer - that's fairly easy to > do. I would have through pacat should be able to do it too tho', but > I've not quite found the right command ;) Since I'm not too familiar with gstreamer, I don't say it definetely won't work, but I'm quite sceptical. The reason I think it probably doesn't work (or maybe works, but badly) is that the "pipeline" is clocked in both ends, and they don't share the clock. For the same reason the pacat solution won't work either. The audio needs to be resampled between the source and the sink. I'm not aware of any solution to this apart from RTP. If RTP isn't used for anything else, the extra network traffic could be prevented with firewalling in the Linux machine. What really is needed is a proper virtual pulse sound card for Windows. -- Tanu Kaskinen
