Notices

Here's what i did, which is a pretty serious hack but it works nicely:

1. I exported the music library on my laptop as an NFS share
2. I installed mpd on the n810, which is available from the garage
3. I copied (important point) the tag_cache file from my laptop, which lists the music files in the database
4. I mount the NFS share on the n810 (http://maemo.org/community/wiki/settingupnfs/)
5. Now I can run mpd on the n810 itself, using NFS as the source for all the music, and use sonata as the frontend.

Note the step above where I mention copying the tag_cache file. This is very important, because when you rescan the mpd library, it's going to basically download all of the mp3s over nfs to get the tag data. I tried this, and it does not work. Luckily the tag_cache file is an extremely simple file format, so you can just copy it from the main computer (which also, therefore, needs mpd set up).

With a few scripts to automate nfs mounting, this is an easy way to get all the music from your computer onto your nokia.

Oops, I guess I'll stay with UPnP for this task then.

what do I have to install, so that lyrics work?

Lyrics support in Sonata is ****. I installed ZSI on my desktop with Sonata and every song I have, Sonata keeps saying no lyrics found.

Yeah.... TwonkyVision and Canola2 work very well together.

I was glad to see this thread. I had never heard of mpd. I got it working on my server but so far I haven't gotten any network connected clients to connect to it. I opened port 6600 on the server but so far a Windows PC and my IT cannot connect yet.

Update: I changed the mpd conf file from bind to 127.0.0.1 to "any" and all PCs on the network clients can connect now.


Thanks for the thread,
Frank

Hi ywwg,
well done, now I'm able to control my mpd server flawlessly using my N800.

I've just a couple of observations: the artist/album tags are correctly understood, but no cover art is loaded, despite the tick in the preferences.

Other than that, when found the lyrics are always loaded partially, let's say the first 10/20% of the whole song. Is it possible to have the full text, like with Sonata "desktop"?

Thanks again
Fabrizio

One should be able to use PulseAudio and select a different sink than default (localhost) on the N900's PulseAudio daemon. In GNOME, the PulseAudio applet allows to select different server (the PA daemon to control), sink (ie. target, output), and source (ie. input). UPnP-AV and RTP should work as well. But the issue is then the playlists and database of files. In my case, because I use rhythmbox, I can use SSH client + rhythmbox-client to control.