Re: [sc-dev] Convolution2 virtual room

[James Harkins <jamshark70@xxxxxx>]

On December 12, 2015 4:37:36 AM "Victor Bombi" <sonoro@xxxxxxxxxxxxxx> wrote:

> When room changes continuously (once per 64 samples frame) the trigger dont
> works because it is always 1.0.
>
> I think that would be better that trigger behaviour were as with t_trig, so
> that 1.0 meant always update.
>
> What can be done?

That's a HUGE change in semantics that you're proposing. Triggers have 
always been defined as a positive sample coming after a zero/negative 
sample. I would worry very much about changing this, even in one 
exceptional case.

Currently you can use Trig1 to block triggers for the next "dur" seconds. 
Under your proposal, Trig1 would be interpreted as "continuously update for 
the next 'dur' seconds" -- exactly the opposite behavior! Apart from 
breaking code, this would be inconsistent with the rest of the class library.

My suggestion is to work out a way to alternate 1 and 0 so that you update 
once every 2 control cycles.

I'm fairly sure also that it's meaningless to change the Convolution2 
kernel buffer while it's in the middle of an FFT frame... so the goal of 
changing every control cycle may simply be moot from the start.

hjh

Sent with AquaMail for Android
http://www.aqua-mail.com





_______________________________________________
sc-dev mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: https://listarc.bham.ac.uk/marchives/sc-dev/
search: https://listarc.bham.ac.uk/lists/sc-dev/search/