User model/expectations of software synths.
User model/expectations of software synths.
- Subject: User model/expectations of software synths.
- From: Ian Kemmish <email@hidden>
- Date: Fri, 8 May 2009 20:50:17 +0100
Having gone to the effort of writing a completely multi-timbral
software synth, you can imagine my disappointment at learning that the
commonest scenario is that the host app creates one instance of the AU
for each track on which it is used, and always talks to it on Group
0 :-)
For a long time, the only reason this bothered me was because it would
lead to increased cache thrashing during rendering, something I've
expended elbow grease on avoiding.
But I've just had another thought, about what behaviour the user
expects. Supposing the user edits a voice in one particular
instance's user interface - and by editing a voice here I specifically
mean changing a property, not changing a parameter (for example,
adding a module to a voice in a modular synth). Does he expect that
edit to be reflected in all the other instances immediately, or the
next time he saves the edited voice, or the next time he starts the
host app?
I can think of arguments in favour of all of these, but at the moment
I'm leaning towards "immediately" which I guess requires some kind of
shared voice memory between different instances of the same AU.
Does Apple have any official guidance on this? It's possible the AU
Preset mechanism already has stuff to handle this, but I don't recall
reading anything about it in the documentation.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - -
Ian Kemmish 18 Durham Close, Biggleswade, Beds
SG18 8HZ
email@hidden Tel: +44 1767 601361 Mob: +44 7952
854387
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - -
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden