Michaelwagner;136910 Wrote: > > There are several sources for gaps. Indeed.
> > The gap could be on the original recording. Despite that, some people > want to get rid of it. > Forget it - silence detection is a pain in the butt and not worth the effort. I don't know anyone personally who wants that kind of gapless support - we're not talking about removing gaps, we're talking about _not_adding_ them. > > The gap could have been encoded by the mp3 encoder as extra silence > (some early encoders weren't careful in this area and added their own > ideas about gaps). > Such as basically all mp3 encoders except lame, and things like wma encoders. Again, forget it. If the format doesn't support it (vorbis, flac, aac, wav, etc all do) or if there isn't a definitive simple workaround (lame mp3) then don't support it. > > Some gap can be caused by delays in the server as the next track > starts. I know the server stops serving and pretty much drains the > buffer at the end of each track and then starts again. If the network > is at that moment having problems, the gap can be longer. This is > especially true in wireless. > Slimserver/squeezebox is perfectly capable of gapless playback with very high bitrate files (wav, flac, etc) and does so flawlessly for me on both wired and wireless. This is nothing to worry about. Again, we're not asking for anything which hasn't already been done for other formats. > > It's possible that the player adds gap of it's own. I don't remember if > this question got answered or not. > It doesn't. > > Some gap is added by the player synchronization feature and can only be > removed by another enhancement request, that is, the network clock > synchronization feature, itself a can of worms. > Yes. Sync disables gapless for all formats. I'd like gapless & sync, but that's a whole other discussion. This thread is about support for lame mp3 gapless. > > I would be interested to know how this is solved in FLAC (if in fact it > is solved). > It is, and it's not (conceptually) hard. You just do a little prebuffering and don't stop the decoder. I'll re-iterate this point because it bears repeating - slimserver/squeezebox is already 100% gapless by design for all formats which are themselves gapless. Mp3 is _not_ a gapless format, and in the general case (looking only at the basic mp3 spec) will never be. But for lame, and only lame, a simple workaround for gapless exists. We're asking that it be used in slimserver/squeezebox as it is in _many_ other products including, as of recently, the iPod. > > Some of the competitors who now claim to support gapless do it by > cheating, that is, by merely truncating silence in a fixed way. Since > the definition of silence is artificially imposed rather than a > reproduction of the recording, the playback will still be in some way > incorrect. I'm sure Slim doesn't want to cheat and get it wrong. > No-one wants silence removal. It sucks [as I already mentioned :) ]. Who are these competitors who claim to be gapless but only use silence removal? I'm unaware of any. > > The fix needs both client (that is, SB) and server changes. > Why? > > The player now sees none of the ID3 tags (according to Vidur), so > there's no easy way to just pass the LAME gapless tags through to the > player. Thus new architecture needs to be done between the player and > the server ... something that needs to be done with some care. > IIRC the gapless tags are not ID3, they're some other part of the mp3 spec (I forget the details but I'm sure they're easily discoverable). However, they may still need to be propagated to the client, so you're right, that might need a little work. On the other hand, assuming the server merely strips the ID3 block and passes the actual mp3 structure un-messed-with, they may already be there. My (uneducated) guess is that they're probably already being sent. -- radish ------------------------------------------------------------------------ radish's Profile: http://forums.slimdevices.com/member.php?userid=77 View this thread: http://forums.slimdevices.com/showthread.php?t=27384 _______________________________________________ discuss mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/discuss
