Le Sunday 15 June 2008 16:09:53 David Baelde, vous avez écrit : > Nevertheless I listened to it, and observed the logs. I found a minor > bug. When a sequencing transition starts (i.e. no transition) it is > usually followed by another one 15 seconds later. It's not something > one car hear, I think, but it's stupid. I fixed it. > > Except for that I didn't hear any bug. But I did hear some bad choices > in the transitions. I had a track with almost 2 seconds of silence at > the end, followed by a loudly-starting rock. The result was: > > 2008/06/15 16:01:26 [src_8170:3] Analysis: -141.864481dB / > -12.537607dB (15.01s / 2.02s) > 2008/06/15 16:01:26 [smart_crossfade:3] No transition: just sequencing. > > Indeed, if you look at the code of smart_crossfade in utils.liq, > you'll see that the crossing transitions are not used unless the > levels of the two tracks are close enough, which is not the case here.
I've proposed an enhancement to smart_crossfade: http://savonet.rastageeks.org/attachment/ticket/152 Romain ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php _______________________________________________ Savonet-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/savonet-users
