Hi James!

Le 6 mars 2012 17:06, James Moon <[email protected]> a écrit :
> I'm having a problem with transition on input.harbor.
> Blow is the portion of the code:
>
> # transition function
> def transition(a,b) =
>   log("transition called...")
>   add(normalize=false,
>       [fade.initial(duration=5.,b),fade.final(duration=5.,a)])
> end
>
>
> # how I called switch function
> dj_live = input.harbor(id="live_dj_harbor", dj_harbor_input_mount_point,
> port=dj_harbor_input_port, auth=check_dj_client, buffer=10., max=20.)
> ignore(output.dummy(dj_live, fallible=true))
> switch(id="live_dj_switch", track_sensitive=false, transitions=[transition,
> transition], [({!live_dj_enabled},dj_live), ({true}, s)])
>
> When I turn on/off live_dj_enabled predicate, transition works, but not
> nicely.
>
> When transition from s->dj_live, I notice that s is not fading out
> gradually. It suddenly reduces volume of s and the fade in of dj_live come
> in ok.
> When transition from dj_live->s, samething happens but the only difference
> is that when the volume of old source(dj_live in this case), drop more
> dramatically.
>
> Do you have any idea why I don't get smooth transition?
> Thank you for your support.

As a general advice, you should always remember that input.harbor and
other live sources have to work at real-time. This means that all
transitions that consume data at a higher rate, such as crossfades,
are bound to fail with those.

That being said, I am not sure that this is the case with you. fade.*
operations should not consume more than real-time. However, add() may.

Have you tested with a sequence instead of an add(), just to see what it does?

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

Reply via email to