I think the simplest definition of "large collection" would have to a collection large enough so that the speed tips I suggested would be useful. It depends upon your file system and your use patterns. I have fewer than 4000 tracks in my collection and some things I tried to do were painfully slow, such as editting tags IIRC. When I switched to MySQL I noticed a HUGE improvement. But that was a while ago and it could be that the SQLite performance has improved in the meantime.Pithlit wrote:Define "large collection"... my amarok works just dandy with 5k+ songs. Without a db backend and with watching the collection for new stuff.
I've tried both mysql and sqlite in recently, and I haven't noticed any speed differences when searching through the collection (I have about 25000 songs).BitJam wrote:I think the simplest definition of "large collection" would have to a collection large enough so that the speed tips I suggested would be useful. It depends upon your file system and your use patterns. I have fewer than 4000 tracks in my collection and some things I tried to do were painfully slow, such as editting tags IIRC. When I switched to MySQL I noticed a HUGE improvement. But that was a while ago and it could be that the SQLite performance has improved in the meantime.Pithlit wrote:Define "large collection"... my amarok works just dandy with 5k+ songs. Without a db backend and with watching the collection for new stuff.
The suggestions I gave resulted from my own personal experience and also from complaints and solutions I've seen in these forums. Since the OP is shopping around for a music app to handle his "nearly unmanageably large music collection" I thought it would be beneficial to give him some clues to show amaroK in its best (fastest) light so he can make an informed decision.
Speaking of which, another tip for using amaroK is to use the dynamic playlist feature if you want to listen to random selections from all or parts of your large collection. This performs much better than creating a very large playlist.
I'd have to argue against MPD. The last time I used MPD, the sound quality was much worse than any other program I used to play music. If I recall, everything sounded "ting"-y. I think it had to do with my sound card and drivers, as I have a Soundblaster Live 24-bit. I don't think I had this problem with my old Soundblaster Live! 5.1. But who knows, maybe it's time to try mpd again. Hopefully it's less of a pain in the ass to set up, too.yngwin wrote:Amarok is nice, but can be very slow. Also, I haven't found a way to make it do real gapless playback.
That's where I switched to mpd. As you are on gnome, a client like gmpc will probably provide what you need. MPD is a lot faster than Amarok, and also isn't affected by X problems (Amarok would have 'hiccups' under high X loads, and even when X crashes mpd just plays on).
mpd also creates a db, also manages playlists (and using some clients you will be able to fetch album art, lyrics, wikipedia, info...). It doesn't 'just play songs' [1].Pithlit wrote:How does mpd help managing a (nearly unmanagable) collection? I always thought mpd just plays songs...
Depends what you mean with managing. MPD organizes things by tags, which is what TS is looking for. If you actually need to edit tags, I'd suggest Easytag. But MPD handles large playlists better than anything else I know.Pithlit wrote:How does mpd help managing a (nearly unmanagable) collection? I always thought mpd just plays songs...
It works fine here. I use the Xine backend with xine-lib-1.1.8. Live albums play seamlessly.yngwin wrote:Amarok can indeed do gapless playback in theory. But in practice it doesn't have a backend where this really works.
MPD plays them flawlessly. Amarok with xine backend didn't. It may be that they finally fixed this is latest xine.BitJam wrote:Maybe there are artifacts at the end/start of your music files that make you think there is a gap.
(Another reason is that due to continued exposure to an IT major at a technical university, I have developed a pathological phobia of unnecessary redundancy.)
