I've search on Google as well as these forums and found nothing about Amarok supporting a "Play To" mode where by the audio is sent to a DLNA Play To compliant device, like in WMP (not that I've ever used that). I use Amarok exclusively for listening and collection management, so it would be cool if I could create a playlist and then have it play to my audio receiver.

If this already exists, someone please tell me how to enable it. If it's already been proposed, add a "second" to the proposal.

You can probably accomplish this through PulseAudio.

I don't use DLNA, but I have PulseAudio setup to pipe music from my server to speakers in the kitchen (local sound device on server), living room and bedroom (two separate audio cards on a machine in my living room) or my study (a third machine), as well as to one of my laptops wherever it might be (e.g., outdoor on the patio). And at times, I cluster several of them together, so I have the same music playing in the kitchen, study and living room, e.g., when I'm cooking.

Amarok is none the wiser that all of this is going on behind its back, and it works with all the music playing apps I've tried, including Grooveshark running in the web browser or Spotify running under WINE.

could you give some details on how you do remote binding with pulseaudio? /mark

To start experimentation, just open the relevant ports on the destination machine, UDP 5353 and TCP 16001/4713 and I believe you need to have at least UDP 5353 open on the source.

Then use the pulsaudio configuration tool on the destination machine to enable network access to local sound devices, and similarly on the source machine to make discoverable network sound devices available locally.

Finally, in the normal gnome sound configuration tool on the source, the network devices will appear and you can just select one of them, and now any audio played on the source machine will go to the selected remote device and play from whatever speakers you have attached to it.

Once all of this is working, you can get fancy and use the command line tools to give you more detailed control over things.

re: ports - check.

re: pulseaudio config tool - is called?

re: gnome sound config & sink discovery - this presumes some kind of automatic binding which almost surely requires a common subnet, no?

mtk wrote:re: ports - check.

re: pulseaudio config tool - is called?

re: gnome sound config & sink discovery - this presumes some kind of automatic binding which almost surely requires a common subnet, no?

The one you are looking for is called "paprefs".

And for the simple setup I sketched, yes, you would be using avahi to do service discovery on a local LAN. It's possible to do point-to-point as well, but for that you'll need to consult with someone smarter than me.

pulse audio is usually configured using paprefs; which is a gtk app.

Do some internet searching on enablin upnp and airtunes on pulse and you'll find the tools. Basically you'll need some of the additional pulse backend modules, and a couple of the front end configuration tools (on both machines.) Then you need to have the server/sink (receiver) enable it's sound devices as network devices, and have the source (broadcaster) enabled network devices as sound devices. The rest kind of just pops up.
When you have the devices all enabled, you can use the phonon backend to select devices, but I've found this to be unreliable compared to the gnome frontend (padevchooser.)

The package names are distro specific, but it's usually easy to find the help online

Incidentally, in my experience, I've found the sound quality to be high, but the service stability to be low. Other people have had reports of great setups though.