I've always had conservative=true. This doesn't seem to have an impact apparently.
I'm not using the new cue_cut operator in my script, either (although I'm really tempted to ;). Curiously, when is the fade-out supposed to start? Does it depend on the start_next parameter or rather on the duration of the fade-out, e.g. counting backwards from the track ending? Henry On 16/11/2011 01:35, Romain Beauxis wrote: > Hi, > > 2011/11/15 Henry Favretto <[email protected]>: >> Actually, I did some more testing and increased the start_next parameter to >> 200 seconds just to >> provoke an early crossfade and I noticed that the fade-out of the first >> source simply doesn't happen >> for the default transition e.g. I can hear the second track fade-in but then >> the first track keeps >> playing in the background until the end. Interestingly, for all the other >> transition cases >> (non-default) this seems to work okay. >> >> I just thought I'd let you know since maybe that helps with narrowing down >> the problem. > > Your diagnostic seems correct indeed. > > Just one question: are you using also the new cue_cut operator? > > In any case, you should look at adding conservative=true to your smart > crossfade. This is designed to help by buffering in advance what is > needed for the transition. > > Romain ------------------------------------------------------------------------------ RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Savonet-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/savonet-users
