Hey, 2012/3/9 David Baelde <[email protected]>: > I'd like to clarify one thing: there is never any (clock) problem with > switch-based transitions. It's only cross-transitions (between tracks > of the same source) that you should avoid with real time sources (we > decided against this so far, but strict clock annotations would > prevent such uses). > > Regarding the initial problem, I'm not sure what it was, maybe bad > remaining time estimations. I'm more puzzled by the intermediate > report of liquidsoap going into a loop with a sequence transitions.
It would be nice to investigate on those too indeed. > Finally, the override parameter did not make sense in > fade.initial/final(), only in fade.in/out(). It is there to change the > fading duration from one track to the next. This is useless for > track.initial/final() because they fade only once, not for every > track. (Strictly speaking, it could still be nice to use a duration > extracted from the track, and technically there may be a way that even > initial/final extract it before starting their work, but I'm not > sure.) Your use of references for setting fading durations seems like > the right approach to me. Thanks. It makes sense to not have metadata override for fade.intial/fade.final. However, I think that the problem here is that those fade operators actually are overriden by those metadata.. Romain ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ _______________________________________________ Savonet-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/savonet-users
